You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2022/03/17 02:36:39 UTC
[tvm-site] branch asf-site2 updated: deploying docs (apache/tvm@6eaa5e6f29420b2c85ec6e3c5f83566d60c7d641)
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a commit to branch asf-site2
in repository https://gitbox.apache.org/repos/asf/tvm-site.git
The following commit(s) were added to refs/heads/asf-site2 by this push:
new eaa5ea9 deploying docs (apache/tvm@6eaa5e6f29420b2c85ec6e3c5f83566d60c7d641)
eaa5ea9 is described below
commit eaa5ea9407e5dfa4e54a654c0505d03f13ac6a35
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Thu Mar 17 02:36:33 2022 +0000
deploying docs (apache/tvm@6eaa5e6f29420b2c85ec6e3c5f83566d60c7d641)
---
.../micro_tvmc.ipynb | 12 +-
.../eb483c672b88006c331115968e0ffd9b/micro_tvmc.py | 66 +-
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_paddle.rst.txt | 2 +-
.../how_to/compile_models/from_pytorch.rst.txt | 2 +-
.../how_to/compile_models/from_tensorflow.rst.txt | 2 +-
.../compile_models/sg_execution_times.rst.txt | 22 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 18 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 4 +-
.../how_to/extend_tvm/sg_execution_times.rst.txt | 10 +-
.../how_to/extend_tvm/use_pass_instrument.rst.txt | 16 +-
.../optimize_operators/opt_conv_cuda.rst.txt | 2 +-
.../optimize_operators/opt_conv_tensorcore.rst.txt | 2 +-
.../how_to/optimize_operators/opt_gemm.rst.txt | 28 +-
.../optimize_operators/sg_execution_times.rst.txt | 8 +-
.../sg_execution_times.rst.txt | 16 +-
.../tune_conv2d_layer_cuda.rst.txt | 3197 +++++++-------------
.../tune_network_cuda.rst.txt | 2 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 49 +-
.../tune_with_autotvm/sg_execution_times.rst.txt | 12 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 34 +-
.../work_with_microtvm/micro_autotune.rst.txt | 16 +-
.../how_to/work_with_microtvm/micro_tflite.rst.txt | 2 +-
.../how_to/work_with_microtvm/micro_tvmc.rst.txt | 66 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 12 +-
.../work_with_relay/sg_execution_times.rst.txt | 8 +-
.../schedule_primitives.rst.txt | 14 +-
.../work_with_schedules/sg_execution_times.rst.txt | 18 +-
.../how_to/work_with_schedules/tensorize.rst.txt | 8 +-
.../work_with_schedules/tuple_inputs.rst.txt | 24 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 6 +-
.../frontend/deploy_classification.rst.txt | 2 +-
.../tutorials/frontend/deploy_detection.rst.txt | 2 +-
.../tutorials/frontend/sg_execution_times.rst.txt | 6 +-
.../vta/tutorials/optimize/convolution_opt.rst.txt | 4 +-
.../tutorials/optimize/matrix_multiply_opt.rst.txt | 4 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 2 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 57 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 26 +-
.../tutorial/tensor_expr_get_started.rst.txt | 61 +-
docs/commit_hash | 2 +-
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_paddle.html | 2 +-
docs/how_to/compile_models/from_pytorch.html | 7 +-
docs/how_to/compile_models/from_tensorflow.html | 2 +-
docs/how_to/compile_models/sg_execution_times.html | 22 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 31 +-
docs/how_to/deploy_models/deploy_prequantized.html | 12 +-
.../deploy_models/deploy_prequantized_tflite.html | 4 +-
docs/how_to/deploy_models/deploy_quantized.html | 2 +-
docs/how_to/deploy_models/deploy_ssd_gluoncv.html | 39 +-
docs/how_to/deploy_models/sg_execution_times.html | 18 +-
.../extend_tvm/bring_your_own_datatypes.html | 4 +-
docs/how_to/extend_tvm/sg_execution_times.html | 10 +-
docs/how_to/extend_tvm/use_pass_instrument.html | 16 +-
docs/how_to/optimize_operators/opt_conv_cuda.html | 2 +-
.../optimize_operators/opt_conv_tensorcore.html | 2 +-
docs/how_to/optimize_operators/opt_gemm.html | 28 +-
.../optimize_operators/sg_execution_times.html | 8 +-
.../sg_execution_times.html | 14 +-
.../tune_conv2d_layer_cuda.html | 3197 +++++++-------------
.../tune_with_autoscheduler/tune_network_cuda.html | 2 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 49 +-
.../tune_with_autotvm/sg_execution_times.html | 12 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 34 +-
docs/how_to/work_with_microtvm/micro_autotune.html | 16 +-
docs/how_to/work_with_microtvm/micro_tflite.html | 2 +-
docs/how_to/work_with_microtvm/micro_tvmc.html | 51 +-
.../work_with_microtvm/sg_execution_times.html | 12 +-
.../how_to/work_with_relay/sg_execution_times.html | 8 +-
.../work_with_schedules/schedule_primitives.html | 14 +-
.../work_with_schedules/sg_execution_times.html | 18 +-
docs/how_to/work_with_schedules/tensorize.html | 8 +-
docs/how_to/work_with_schedules/tuple_inputs.html | 24 +-
docs/reference/api/python/auto_scheduler.html | 4 +-
docs/reference/api/python/target.html | 3 +
.../api/typedoc/classes/bytestreamreader.html | 12 +-
.../api/typedoc/classes/cachedcallstack.html | 34 +-
docs/reference/api/typedoc/classes/dldatatype.html | 12 +-
docs/reference/api/typedoc/classes/dldevice.html | 10 +-
.../reference/api/typedoc/classes/environment.html | 12 +-
docs/reference/api/typedoc/classes/ffilibrary.html | 20 +-
.../api/typedoc/classes/graphexecutor.html | 16 +-
docs/reference/api/typedoc/classes/instance.html | 40 +-
docs/reference/api/typedoc/classes/memory.html | 34 +-
docs/reference/api/typedoc/classes/module.html | 10 +-
docs/reference/api/typedoc/classes/ndarray.html | 22 +-
.../api/typedoc/classes/packedfunccell.html | 6 +-
docs/reference/api/typedoc/classes/rpcserver.html | 14 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
.../api/typedoc/classes/webgpucontext.html | 12 +-
docs/reference/api/typedoc/enums/argtypecode.html | 30 +-
.../api/typedoc/enums/aynccallbackcode.html | 4 +-
.../api/typedoc/enums/dldatatypecode.html | 8 +-
.../api/typedoc/enums/rpcserverstate.html | 12 +-
docs/reference/api/typedoc/enums/sizeof.html | 18 +-
docs/reference/api/typedoc/index.html | 112 +-
.../api/typedoc/interfaces/disposable.html | 2 +-
.../api/typedoc/interfaces/functioninfo.html | 6 +-
.../api/typedoc/interfaces/libraryprovider.html | 4 +-
docs/searchindex.js | 2 +-
.../vta/tutorials/autotvm/sg_execution_times.html | 6 +-
.../tutorials/frontend/deploy_classification.html | 2 +-
.../vta/tutorials/frontend/deploy_detection.html | 2 +-
.../vta/tutorials/frontend/sg_execution_times.html | 6 +-
.../vta/tutorials/optimize/convolution_opt.html | 4 +-
.../tutorials/optimize/matrix_multiply_opt.html | 4 +-
.../vta/tutorials/optimize/sg_execution_times.html | 6 +-
docs/topic/vta/tutorials/sg_execution_times.html | 6 +-
docs/tutorial/auto_scheduler_matmul_x86.html | 2 +-
docs/tutorial/autotvm_relay_x86.html | 171 +-
docs/tutorial/cross_compilation_and_rpc.html | 2 +-
docs/tutorial/intro_topi.html | 2 +-
docs/tutorial/sg_execution_times.html | 26 +-
docs/tutorial/tensor_expr_get_started.html | 57 +-
128 files changed, 3038 insertions(+), 5289 deletions(-)
diff --git a/docs/_downloads/6e511f5a8ddbf12f2fca2dfadc0cc4a9/micro_tvmc.ipynb b/docs/_downloads/6e511f5a8ddbf12f2fca2dfadc0cc4a9/micro_tvmc.ipynb
index aacbff9..ad280db 100644
--- a/docs/_downloads/6e511f5a8ddbf12f2fca2dfadc0cc4a9/micro_tvmc.ipynb
+++ b/docs/_downloads/6e511f5a8ddbf12f2fca2dfadc0cc4a9/micro_tvmc.ipynb
@@ -29,42 +29,42 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "Using TVMC Micro\n###########################################################\n\n TVMC is a command-line tool which is installed as a part of TVM Python packages. Accessing this\n package varies based on your machine setup. In many cases, you can use the ``tvmc`` command directly.\n Alternatively, if you have TVM as a Python module on your ``$PYTHONPATH``, you can access this\n driver with ``python -m tvm.driver.tvmc`` command. This tutorial will use TVMC command as\n ``tvmc`` f [...]
+ "Using TVMC Micro\n###########################################################\n\n TVMC is a command-line tool which is installed as a part of TVM Python packages. Accessing this\n package varies based on your machine setup. In many cases, you can use the ``tvmc`` command directly.\n Alternatively, if you have TVM as a Python module on your ``$PYTHONPATH``, you can access this\n driver with ``python -m tvm.driver.tvmc`` command. This tutorial will use TVMC command as\n ``tvmc`` f [...]
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "Obtain a Tiny Model\n###########################################################\n\n For this tutorial, we will use Magic Wand model from tflite micro. Magic Wand is a\n Depthwise Convolution Layer model which recognizes gestures with an accelerometer.\n\n For this tutorial we will be using the model in tflite format.\n\n .. code-block:: bash\n\n wget https://github.com/tensorflow/tflite-micro/raw/main/tensorflow/lite/micro/examples/magic_wand/magic_wand.tflite\n\n\n"
+ "Obtain a Tiny Model\n###########################################################\n\n For this tutorial, we will use Magic Wand model from tflite micro. Magic Wand is a\n Depthwise Convolution Layer model which recognizes gestures with an accelerometer.\n\n For this tutorial we will be using the model in tflite format.\n\n .. code-block:: bash\n\n\t wget https://github.com/tensorflow/tflite-micro/raw/main/tensorflow/lite/micro/examples/magic_wand/magic_wand.tflite\n\n\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "Compiling a TFLite model to a Model Library Format\n###########################################################\n\n Model Library Format (MLF) is an output format that TVM provides for micro targets. MLF is a tarball\n containing a file for each piece of the TVM compiler output which can be used on micro targets outside\n TVM environment. Read more about `Model Library Format <https://tvm.apache.org/docs//arch/model_library_format.html>`_.\n\n Here, we generate a MLF file for `` [...]
+ "Compiling a TFLite model to a Model Library Format\n###########################################################\n\n Model Library Format (MLF) is an output format that TVM provides for micro targets. MLF is a tarball\n containing a file for each piece of the TVM compiler output which can be used on micro targets outside\n TVM environment. Read more about `Model Library Format <https://tvm.apache.org/docs//arch/model_library_format.html>`_.\n\n Here, we generate a MLF file for `` [...]
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "Create a Zephyr Project Using Model Library Format\n###########################################################\n\n To generate a Zephyr project we use TVM Micro subcommand ``create``. We pass the MLF format and the path\n for the project to ``create`` subcommand along with project options. Project options for each\n platform (Zephyr/Arduino) are defined in their Project API server file. To generate Zephyr project, run:\n\n .. code-block:: bash\n\n tvmc micro create \\\n [...]
+ "Create a Zephyr Project Using Model Library Format\n###########################################################\n\n To generate a Zephyr project we use TVM Micro subcommand ``create``. We pass the MLF format and the path\n for the project to ``create`` subcommand along with project options. Project options for each\n platform (Zephyr/Arduino) are defined in their Project API server file. To generate Zephyr project, run:\n\n .. code-block:: bash\n\n\t tvmc micro create \\\n\t [...]
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "Build and Flash Zephyr Project Using TVMC Micro\n###########################################################\n\n Next step is to build the Zephyr project which includes TVM generated code for running the tiny model, Zephyr\n template code to run a model in Host-Driven mode and TVM runtime source/header files. To build the project:\n\n .. code-block:: bash\n\n tvmc micro build \\\n project \\\n zephyr \\\n --project-option zephyr_board=qemu_x86\n\n This will [...]
+ "Build and Flash Zephyr Project Using TVMC Micro\n###########################################################\n\n Next step is to build the Zephyr project which includes TVM generated code for running the tiny model, Zephyr\n template code to run a model in Host-Driven mode and TVM runtime source/header files. To build the project:\n\n .. code-block:: bash\n\n\t tvmc micro build \\\n\t project \\\n\t zephyr \\\n\t --project-option zephyr_board=qemu_x86\n\n This wi [...]
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "Run Tiny Model on Micro Target\n###########################################################\n\n After flashing the device, the compiled model and TVM RPC server are programmed on the device.\n The Zephyr board is waiting for host to open a communication channel. MicroTVM devices typicall communicate\n using a serial communication (UART). To run the flashed model on the device using TVMC, we use ``tvmc run`` subcommand\n and pass ``--device micro`` to specify the device type. Thi [...]
+ "Run Tiny Model on Micro Target\n###########################################################\n\n After flashing the device, the compiled model and TVM RPC server are programmed on the device.\n The Zephyr board is waiting for host to open a communication channel. MicroTVM devices typicall communicate\n using a serial communication (UART). To run the flashed model on the device using TVMC, we use ``tvmc run`` subcommand\n and pass ``--device micro`` to specify the device type. Thi [...]
]
}
],
diff --git a/docs/_downloads/eb483c672b88006c331115968e0ffd9b/micro_tvmc.py b/docs/_downloads/eb483c672b88006c331115968e0ffd9b/micro_tvmc.py
index 423e0f1..a439eb3 100644
--- a/docs/_downloads/eb483c672b88006c331115968e0ffd9b/micro_tvmc.py
+++ b/docs/_downloads/eb483c672b88006c331115968e0ffd9b/micro_tvmc.py
@@ -57,14 +57,15 @@ and run the model all using `tvmc micro` command.
#
# .. code-block:: bash
#
-# tvmc --help
+# tvmc --help
#
# To compile a model for microtvm we use ``tvmc compile`` subcommand. The output of this command
# is used in next steps with ``tvmc micro`` subcommands. You can check the availability of TVMC Micro using:
#
# .. code-block:: bash
#
-# tvmc micro --help
+# tvmc micro --help
+#
#
# The main tasks that you can perform using ``tvmc micro`` are ``create``, ``build`` and ``flash``.
# To read about specific options under a givern subcommand, use
@@ -82,7 +83,7 @@ and run the model all using `tvmc micro` command.
#
# .. code-block:: bash
#
-# wget https://github.com/tensorflow/tflite-micro/raw/main/tensorflow/lite/micro/examples/magic_wand/magic_wand.tflite
+# wget https://github.com/tensorflow/tflite-micro/raw/main/tensorflow/lite/micro/examples/magic_wand/magic_wand.tflite
#
############################################################
@@ -97,16 +98,16 @@ and run the model all using `tvmc micro` command.
#
# .. code-block:: bash
#
-# tvmc compile magic_wand.tflite \
-# --target='c -keys=cpu -link-params=0 -model=host' \
-# --runtime=crt \
-# --runtime-crt-system-lib 1 \
-# --executor='graph' \
-# --executor-graph-link-params 0 \
-# --output model.tar \
-# --output-format mlf \
-# --pass-config tir.disable_vectorize=1 \
-# --disabled-pass=AlterOpLayout
+# tvmc compile magic_wand.tflite \
+# --target='c -keys=cpu -link-params=0 -model=host' \
+# --runtime=crt \
+# --runtime-crt-system-lib 1 \
+# --executor='graph' \
+# --executor-graph-link-params 0 \
+# --output model.tar \
+# --output-format mlf \
+# --pass-config tir.disable_vectorize=1 \
+# --disabled-pass=AlterOpLayout
#
# This will generate a ``model.tar`` file which contains TVM compiler output files. To run this command for
# a different Zephyr device, you need to update ``target``. For instance, for ``nrf5340dk_nrf5340_cpuapp`` board
@@ -124,11 +125,11 @@ and run the model all using `tvmc micro` command.
#
# .. code-block:: bash
#
-# tvmc micro create \
-# project \
-# model.tar \
-# zephyr \
-# --project-option project_type=host_driven zephyr_board=qemu_x86
+# tvmc micro create \
+# project \
+# model.tar \
+# zephyr \
+# --project-option project_type=host_driven zephyr_board=qemu_x86
#
# This will generate a ``Host-Driven`` Zephyr project for ``qemu_x86`` Zephyr board. In Host-Driven template project,
# the Graph Executor will run on host and perform the model execution on Zephyr device by issuing commands to the
@@ -150,10 +151,10 @@ and run the model all using `tvmc micro` command.
#
# .. code-block:: bash
#
-# tvmc micro build \
-# project \
-# zephyr \
-# --project-option zephyr_board=qemu_x86
+# tvmc micro build \
+# project \
+# zephyr \
+# --project-option zephyr_board=qemu_x86
#
# This will build the project in ``project`` directory and generates binary files under ``project/build``. To build
# Zephyr project for a different Zephyr board, change ``zephyr_board`` project option.
@@ -163,10 +164,10 @@ and run the model all using `tvmc micro` command.
#
# .. code-block:: bash
#
-# tvmc micro flash \
-# project \
-# zephyr \
-# --project-option zephyr_board=qemu_x86
+# tvmc micro flash \
+# project \
+# zephyr \
+# --project-option zephyr_board=qemu_x86
#
############################################################
@@ -181,12 +182,13 @@ and run the model all using `tvmc micro` command.
#
# .. code-block:: bash
#
-# tvmc run \
-# --device micro \
-# project \
-# --project-option zephyr_board=qemu_x86 \
-# --fill-mode ones
-# --print-top 4
+# tvmc run \
+# --device micro \
+# project \
+# --project-option zephyr_board=qemu_x86 \
+# --fill-mode ones \
+# --print-top 4
+#
# # Output:
# #
# # INFO:__main__:b'[100%] [QEMU] CPU: qemu32,+nx,+pae\n'
diff --git a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
index 97620bd..3d43f29 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -98,7 +98,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip1e1ff422-41ea-4070-b3fe-e7ac181ee7fa from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip248dc537-6341-4c0b-a084-22d8eb23d2ad from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
x (1, 3, 224, 224)
diff --git a/docs/_sources/how_to/compile_models/from_paddle.rst.txt b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
index e8d246c..0d9fdaa 100644
--- a/docs/_sources/how_to/compile_models/from_paddle.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
@@ -173,7 +173,7 @@ Look up prediction top 1 index in 1000 class synset.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 5.840 seconds)
+ **Total running time of the script:** ( 1 minutes 3.428 seconds)
.. _sphx_glr_download_how_to_compile_models_from_paddle.py:
diff --git a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
index 332b3fe..892e004 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -79,7 +79,7 @@ Load a pretrained PyTorch model
.. code-block:: none
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
0%| | 0.00/44.7M [00:00<?, ?B/s]
5%|5 | 2.38M/44.7M [00:00<00:01, 24.9MB/s]
15%|#4 | 6.66M/44.7M [00:00<00:01, 36.7MB/s]
45%|####4 | 20.1M/44.7M [00:00<00:00, 84.2MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 117MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
29%|##9 | 13.1M/44.7M [00:00<00:00, 137MB/s]
69%|######9 | 31.0M/44.7M [00:00<00:00, 167MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 173MB/s]
diff --git a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
index 91ba962..9e34aca 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -380,7 +380,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 4.056 seconds)
+ **Total running time of the script:** ( 1 minutes 7.096 seconds)
.. _sphx_glr_download_how_to_compile_models_from_tensorflow.py:
diff --git a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
index ff27107..6dc4062 100644
--- a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
@@ -5,15 +5,15 @@
Computation times
=================
-**05:29.039** total execution time for **how_to_compile_models** files:
+**05:28.088** total execution time for **how_to_compile_models** files:
-- **01:05.840**: :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)
-- **01:04.056**: :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``)
-- **00:56.106**: :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)
-- **00:41.382**: :ref:`sphx_glr_how_to_compile_models_from_caffe2.py` (``from_caffe2.py``)
-- **00:25.177**: :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)
-- **00:20.400**: :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)
-- **00:20.264**: :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)
-- **00:19.275**: :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)
-- **00:14.105**: :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)
-- **00:02.433**: :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)
+- **01:07.096**: :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``)
+- **01:03.428**: :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)
+- **00:54.993**: :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)
+- **00:41.121**: :ref:`sphx_glr_how_to_compile_models_from_caffe2.py` (``from_caffe2.py``)
+- **00:25.700**: :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)
+- **00:21.032**: :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)
+- **00:20.145**: :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)
+- **00:18.803**: :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)
+- **00:13.426**: :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)
+- **00:02.343**: :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
index 99e44fb..acb1ba7 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
@@ -393,7 +393,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.1042 16.0792 16.7479 15.5151 0.4439
+ 15.5896 15.6136 15.6952 15.4707 0.0816
diff --git a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
index 017535e..b02ed63 100644
--- a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
@@ -108,7 +108,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
.. code-block:: none
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
0%| | 0.00/170M [00:00<?, ?B/s]
11%|#1 | 19.1M/170M [00:00<00:00, 200MB/s]
26%|##6 | 44.2M/170M [00:00<00:00, 237MB/s]
41%|#### | 69.4M/170M [00:00<00:00, 249MB/s]
56%|#####6 | 95.3M/170M [00:00<00:00, 258MB/s]
71%|#######1 | 121M/170M [00:00<00:00, 262MB/s]
86%|########6 | 146M/170M [00:00<00:00, 264MB/s]
100%|##########| 170M/170M [00:00<00:00, 257MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
2%|2 | 3.45M/170M [00:00<00:04, 35.9MB/s]
5%|5 | 9.11M/170M [00:00<00:03, 49.3MB/s]
11%|# | 18.2M/170M [00:00<00:02, 69.9MB/s]
15%|#4 | 24.9M/170M [00:00<00:02, 54.5MB/s]
18%|#7 | 30.5M/170M [00:00<00:02, 50.4MB/s]
21%|## | 35.5M/170M [00:00<00:03, 40.6MB/s]
26%|##5 | 43.7M/170M [00:00<00:02, 51.2MB/s]
29%|##9 | 49.8M/170M [00:01<00:02, 54.1MB/s]
33%|###2 | 55.4M/170M [00:01<00:03, 36.3MB/s]
36%|###6 | 61.7M/170M [00:01<00:02, 42.1MB/s]
40%|#### | 68.5M/170M [00:01<00:02, 48.8MB/s]
44%|####3 | 74.1M/170M [00:01<00:02, 49.0MB/s]
50%|####9 | 84.1M/170M [00:01<00:01, 62.3MB/s]
56%|#####6 | 95.2M/170M [00:01<00:01, 76.5MB/s]
64%|######3 | 108M/170M [00:01<00:00, 92.0MB/s]
69%|######9 | 117M/170M [00:02<00:00, 93.5MB/s]
76%|#######5 | 128M/170M [00:02<00:00, 99.9MB/s]
82%|########1 | 139M/170M [00:02<00:00, 102MB/s]
88%|########7 | 149M/170M [00:02<00:00, 85.6MB/s]
93%|#########3| 158M/170M [00:02<00:00, 47.0MB/s]
97%|#########6| 165M/170M [00:02<00:00, 50.6MB/s]
100%|##########| 170M/170M [00:03<00:00, 59.3MB/s]
/usr/local/lib/python3.6/dist-packages/torch/functional.py:445: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:2157.)
return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]
/usr/local/lib/python3.6/dist-packages/torch/nn/functional.py:3702: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
@@ -255,7 +255,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 59.751 seconds)
+ **Total running time of the script:** ( 2 minutes 57.698 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_object_detection_pytorch.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
index eef9377..5209ce8 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -187,7 +187,7 @@ training. Other models require a full post training calibration.
.. code-block:: none
Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
-
0%| | 0.00/13.6M [00:00<?, ?B/s]
42%|####1 | 5.69M/13.6M [00:00<00:00, 59.6MB/s]
84%|########3 | 11.4M/13.6M [00:00<00:00, 58.7MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 59.9MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
19%|#9 | 2.59M/13.6M [00:00<00:00, 27.1MB/s]
38%|###8 | 5.17M/13.6M [00:00<00:00, 25.7MB/s]
56%|#####6 | 7.64M/13.6M [00:00<00:00, 25.4MB/s]
82%|########1 | 11.1M/13.6M [00:00<00:00, 29.5MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 30.6MB/s]
@@ -348,7 +348,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.2507 90.1289 91.9421 89.9283 0.3267
+ 90.4215 90.5000 92.3945 89.8433 0.3587
@@ -388,7 +388,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 6.844 seconds)
+ **Total running time of the script:** ( 1 minutes 6.068 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_prequantized.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
index 037ae47..48b9a54 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
@@ -352,7 +352,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 117.8534 117.8098 122.0296 116.9189 0.6196
+ 116.9473 116.7776 132.1289 115.4894 1.6453
@@ -386,7 +386,7 @@ Here we give an example of how to measure performance of TVM compiled models.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 59.810 seconds)
+ **Total running time of the script:** ( 1 minutes 59.056 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_prequantized_tflite.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
index a79db99..e001d49 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -221,7 +221,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 13.356 seconds)
+ **Total running time of the script:** ( 1 minutes 10.579 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_quantized.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
index b1acfca..0eae544 100644
--- a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
@@ -137,7 +137,7 @@ Convert and compile model for CPU.
data: None
input_sym_arg_type = in_param.infer_type()[0]
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
-
0%| | 0/132723 [00:00<?, ?KB/s]
5%|5 | 7086/132723 [00:00<00:01, 70852.47KB/s]
12%|#2 | 16073/132723 [00:00<00:01, 82036.16KB/s]
19%|#8 | 24678/132723 [00:00<00:01, 83865.46KB/s]
25%|##4 | 33065/132723 [00:00<00:01, 60847.69KB/s]
30%|### | 40087/132723 [00:00<00:01, 55062.75KB/s]
37%|###6 | 48811/132723 [00:00<00:01, 63566.58KB/s]
42%|####1 | 55724/132723 [00:00<00:01, 51224.69KB/s]
49%|####8 | 64444/132723 [00:01<00:01, 59697.83KB/s]
54%|#####3 | 71167/132723 [00:01<00:01, 50614.43KB/s]
58%|#####7 | 76917/132723 [00:01<00:01, 49747.80KB/s]
62%|######2 | 82353/132723 [00:01<00:01, 45939.26KB/s]
69%|######8 | 91049/132723 [00:01<00:00, 55547.15KB/s]
74%|#######4 | 98288/132723 [00:01<00:00, 50102.14KB/s]
80%|######## | 106818/132723 [00:01<00:00, 58229.84KB/s]
85%|########5 | 113458/132723 [00:02<00:00, 47222.48KB/s]
90%|########9
| 118872/132723 [00:02<00:00, 38675.19KB/s]
96%|#########6| 127652/132723 [00:02<00:00, 48387.95KB/s]
100%|##########| 132723/132723 [00:02<00:00, 53825.86KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
5%|4 | 6078/132723 [00:00<00:02, 60773.25KB/s]
9%|9 | 12156/132723 [00:00<00:02, 58694.72KB/s]
14%|#3 | 18031/132723 [00:00<00:02, 56270.28KB/s]
19%|#9 | 25568/132723 [00:00<00:01, 63523.50KB/s]
26%|##6 | 34901/132723 [00:00<00:01, 59493.23KB/s]
31%|###1 | 41301/132723 [00:00<00:01, 60628.27KB/s]
37%|###7 | 49140/132723 [00:00<00:01, 58636.55KB/s]
42%|####1 | 55094/132723 [00:00<00:01, 58796.44KB/s]
47%|####7 | 62859/132723 [00:01<00:01, 64039.76KB/s]
52%|#####2 | 69372/132723 [00:01<00:01, 55440.98KB/s]
57%|#####7 | 75840/132723 [00:01<00:00, 57835.80KB/s]
62%|######1 | 81905/132723 [00:01<00:01, 49313.71KB/s]
68%|######8 | 90545/132723 [00:01<00:00, 58407.61KB/s]
74%|#######3 | 98063/132723 [00:01<00:00, 62756.86KB/s]
79%|#######8 | 104719/132723 [00:01<00:00, 51719.67KB/s]
86%|########6
| 114680/132723 [00:01<00:00, 61724.95KB/s]
94%|#########3| 124198/132723 [00:02<00:00, 70079.29KB/s]
99%|#########9| 131771/132723 [00:02<00:00, 64713.43KB/s]
100%|##########| 132723/132723 [00:02<00:00, 60259.54KB/s]
@@ -202,7 +202,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 24.540 seconds)
+ **Total running time of the script:** ( 2 minutes 17.502 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_ssd_gluoncv.py:
diff --git a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
index 0472867..b3b25c4 100644
--- a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
@@ -5,13 +5,13 @@
Computation times
=================
-**10:33.678** total execution time for **how_to_deploy_models** files:
+**10:19.711** total execution time for **how_to_deploy_models** files:
-- **02:59.751**: :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``)
-- **02:24.540**: :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)
-- **01:59.810**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)
-- **01:13.356**: :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)
-- **01:06.844**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)
-- **00:27.254**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)
-- **00:21.924**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)
-- **00:00.200**: :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``)
+- **02:57.698**: :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``)
+- **02:17.502**: :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)
+- **01:59.056**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)
+- **01:10.579**: :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)
+- **01:06.068**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)
+- **00:26.860**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)
+- **00:21.779**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)
+- **00:00.169**: :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``)
diff --git a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
index eb773a0..7b35e1c 100644
--- a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
@@ -423,7 +423,7 @@ First let us define two helper functions to get the mobilenet model and a cat im
.. code-block:: none
- Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip0246fe14-6daf-4336-90cf-efe6bfe45a86 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipecb261bd-611f-4adb-944c-58ada011921f from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
@@ -525,7 +525,7 @@ Now, to actually convert the entire network, we have written `a pass in Relay <h
.. code-block:: none
- Check failed: (lower) is false: Intrinsic lowering function for target llvm, intrinsic name tir.sqrt, type 150 not found
+ Check failed: (lower) is false: FloatImm lowering function for target llvm type 150 not found
diff --git a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
index 042b365..326023a 100644
--- a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
@@ -5,9 +5,9 @@
Computation times
=================
-**00:38.569** total execution time for **how_to_extend_tvm** files:
+**00:37.579** total execution time for **how_to_extend_tvm** files:
-- **00:34.989**: :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``)
-- **00:02.285**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)
-- **00:01.089**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)
-- **00:00.206**: :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)
+- **00:33.921**: :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``)
+- **00:02.219**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)
+- **00:01.264**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)
+- **00:00.175**: :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)
diff --git a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
index 90bdc3a..7013ef2 100644
--- a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
@@ -199,10 +199,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 6242us [6242us] (45.59%; 45.59%)
- FoldScaleAxis: 7449us [2us] (54.41%; 54.41%)
- FoldConstant: 7447us [1533us] (54.39%; 99.97%)
- InferType: 5914us [5914us] (43.19%; 79.41%)
+ InferType: 6132us [6132us] (45.54%; 45.54%)
+ FoldScaleAxis: 7333us [2us] (54.46%; 54.46%)
+ FoldConstant: 7330us [1513us] (54.44%; 99.97%)
+ InferType: 5818us [5818us] (43.21%; 79.37%)
@@ -239,10 +239,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 6012us [6012us] (44.82%; 44.82%)
- FoldScaleAxis: 7401us [2us] (55.18%; 55.18%)
- FoldConstant: 7399us [1538us] (55.16%; 99.97%)
- InferType: 5861us [5861us] (43.70%; 79.22%)
+ InferType: 5895us [5895us] (44.68%; 44.68%)
+ FoldScaleAxis: 7298us [2us] (55.32%; 55.32%)
+ FoldConstant: 7296us [1532us] (55.30%; 99.97%)
+ InferType: 5765us [5765us] (43.69%; 79.00%)
diff --git a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
index ef09cad..ae31ca6 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -296,7 +296,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 37.764379 ms
+ Convolution: 44.771740 ms
diff --git a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
index 79cf52b..65338ef 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
@@ -627,7 +627,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 7.340276 ms
+ conv2d with tensor core: 7.085309 ms
diff --git a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
index 3487363..f561ecc 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -118,8 +118,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.019323
- Baseline: 3.247969
+ Numpy running time: 0.017687
+ Baseline: 3.448237
@@ -209,7 +209,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.333472
+ Opt1: 0.291418
@@ -307,7 +307,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.346588
+ Opt2: 0.323982
@@ -331,8 +331,8 @@ Here is the generated IR after vectorization.
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
- A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
B: Buffer(B_2: Pointer(float32), float32, [1048576], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (m.outer: int32, 0, 32) {
@@ -398,7 +398,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.116174
+ Opt3: 0.116501
@@ -422,8 +422,8 @@ Here is the generated IR after loop permutation.
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
- A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
B: Buffer(B_2: Pointer(float32), float32, [1048576], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (m.outer: int32, 0, 32) {
@@ -516,7 +516,7 @@ flattening.
.. code-block:: none
- Opt4: 0.110871
+ Opt4: 0.110649
@@ -633,7 +633,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.111154
+ Opt5: 0.111518
@@ -753,7 +753,7 @@ Futhermore, we can also utilize multi-core processors to do the thread-level par
.. code-block:: none
- Opt6: 0.144823
+ Opt6: 0.144946
@@ -777,8 +777,8 @@ Here is the generated IR after parallelization.
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
- C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
+ A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
B: Buffer(B_2: Pointer(float32), float32, [1048576], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
allocate(packedB: Pointer(global float32x32), float32x32, [32768]), storage_scope = global {
diff --git a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
index ce688ef..89524f9 100644
--- a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
@@ -5,8 +5,8 @@
Computation times
=================
-**00:35.235** total execution time for **how_to_optimize_operators** files:
+**00:34.744** total execution time for **how_to_optimize_operators** files:
-- **00:32.549**: :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)
-- **00:01.478**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``)
-- **00:01.208**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)
+- **00:32.154**: :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)
+- **00:01.392**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``)
+- **00:01.199**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
index 5cd5d43..99d93d3 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
@@ -5,11 +5,11 @@
Computation times
=================
-**05:01.378** total execution time for **how_to_tune_with_autoscheduler** files:
-
-- **02:26.864**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``)
-- **01:20.793**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)
-- **00:39.472**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)
-- **00:16.980**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)
-- **00:08.855**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)
-- **00:08.413**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)
+**04:59.143** total execution time for **how_to_tune_with_autoscheduler** files:
+
+- **02:25.965**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``)
+- **01:18.612**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)
+- **00:38.791**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)
+- **00:18.642**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)
+- **00:08.860**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)
+- **00:08.274**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
index d61fba9..e1b2b73 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
@@ -221,1106 +221,543 @@ cooperative fetching, unrolling and operator fusion.
data: Buffer(data_2: Pointer(float32), float32, [25088], []),
kernel: Buffer(kernel_2: Pointer(float32), float32, [2359296], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 32;
- allocate(conv2d_nchw: Pointer(local float32), float32, [7]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [1296]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [2304]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [7], [], scope="local", align=16)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [16]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [2016]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [16], [], scope="local", align=64)[0] = 0f32
conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[2] = 0f32
conv2d_nchw_1[3] = 0f32
conv2d_nchw_1[4] = 0f32
conv2d_nchw_1[5] = 0f32
conv2d_nchw_1[6] = 0f32
- for (rc.outer.outer: int32, 0, 32) {
- let cse_var_2: int32 = (rc.outer.outer*784)
- let cse_var_1: int32 = (rc.outer.outer*144)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [1296], [], scope="shared")[(threadIdx.x_1*6)] = @tir.if_then_else(((((3 <= floormod((threadIdx.x_1*2), 27)) && (floormod((threadIdx.x_1*6), 81) < 72)) && (1 <= floormod((threadIdx.x_1*6), 9))) && (floormod((threadIdx.x_1*6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1*2), 27)*49)) + (floordiv(floormod((threadIdx.x_1*2), 27), 3)*7)) + floormod((threadIdx.x_1*6), 9)) - 8)], 0f32, dtype=float32)
- pad_temp.shared_1[((threadIdx.x_1*6) + 1)] = @tir.if_then_else(((((3 <= floormod((threadIdx.x_1*2), 27)) && (floormod(((threadIdx.x_1*6) + 1), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 1), 9))) && (floormod(((threadIdx.x_1*6) + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1*2), 27)*49)) + (floordiv(floormod((threadIdx.x_1*2), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 1), 9)) - 8)], 0f32, dtype=float32)
- pad_temp.shared_1[((threadIdx.x_1*6) + 2)] = @tir.if_then_else(((((3 <= floormod((threadIdx.x_1*2), 27)) && (floormod(((threadIdx.x_1*6) + 2), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 2), 9))) && (floormod(((threadIdx.x_1*6) + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1*2), 27)*49)) + (floordiv(floormod((threadIdx.x_1*2), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 2), 9)) - 8)], 0f32, dtype=float32)
- pad_temp.shared_1[((threadIdx.x_1*6) + 3)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 1), 27)) && (floormod(((threadIdx.x_1*6) + 3), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 3), 9))) && (floormod(((threadIdx.x_1*6) + 3), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 1), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 1), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 3), 9)) - 8)], 0f32, dtype=float32)
- pad_temp.shared_1[((threadIdx.x_1*6) + 4)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 1), 27)) && (floormod(((threadIdx.x_1*6) + 4), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 4), 9))) && (floormod(((threadIdx.x_1*6) + 4), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 1), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 1), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 4), 9)) - 8)], 0f32, dtype=float32)
- pad_temp.shared_1[((threadIdx.x_1*6) + 5)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 1), 27)) && (floormod(((threadIdx.x_1*6) + 5), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 5), 9))) && (floormod(((threadIdx.x_1*6) + 5), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 1), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 1), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 5), 9)) - 8)], 0f32, dtype=float32)
- }
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 672)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 224), 27)) && (floormod(((threadIdx.x_1*6) + 24), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 6), 9))) && (floormod(((threadIdx.x_1*6) + 6), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 224), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 224), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 6), 9)) - 8)], 0f32, dtype=float32)
- }
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 673)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 224), 27)) && (floormod(((threadIdx.x_1*6) + 25), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 7), 9))) && (floormod(((threadIdx.x_1*6) + 7), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 224), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 224), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 7), 9)) - 8)], 0f32, dtype=float32)
- }
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 674)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 224), 27)) && (floormod(((threadIdx.x_1*6) + 26), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 8), 9))) && (floormod(((threadIdx.x_1*6) + 8), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 224), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 224), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 8), 9)) - 8)], 0f32, dtype=float32)
+ conv2d_nchw_1[7] = 0f32
+ conv2d_nchw_1[8] = 0f32
+ conv2d_nchw_1[9] = 0f32
+ conv2d_nchw_1[10] = 0f32
+ conv2d_nchw_1[11] = 0f32
+ conv2d_nchw_1[12] = 0f32
+ conv2d_nchw_1[13] = 0f32
+ conv2d_nchw_1[14] = 0f32
+ conv2d_nchw_1[15] = 0f32
+ for (rc.outer.outer: int32, 0, 16) {
+ for (ry.outer.outer: int32, 0, 3) {
+ let cse_var_4: int32 = (rc.outer.outer*1568)
+ let cse_var_3: int32 = (ry.outer.outer*7)
+ let cse_var_2: int32 = (rc.outer.outer*288)
+ let cse_var_1: int32 = (ry.outer.outer*3)
+ {
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [2016], [], scope="shared")[(threadIdx.x_1*12)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1*12), 9))) && (floormod((threadIdx.x_1*12), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod((threadIdx.x_1*12), 9)) - 8)], 0f32, dtyp [...]
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 1), 9))) && (floormod(((threadIdx.x_1*12) + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 1), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 2)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 2), 9))) && (floormod(((threadIdx.x_1*12) + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 2), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 3)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 3), 9))) && (floormod(((threadIdx.x_1*12) + 3), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 1), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 3), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 4)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 4), 9))) && (floormod(((threadIdx.x_1*12) + 4), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 1), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 4), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 5)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 5), 9))) && (floormod(((threadIdx.x_1*12) + 5), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 1), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 5), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 6)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 6), 9))) && (floormod(((threadIdx.x_1*12) + 6), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 2), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 6), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 7)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 7), 9))) && (floormod(((threadIdx.x_1*12) + 7), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 2), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 7), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 8)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 8), 9))) && (floormod(((threadIdx.x_1*12) + 8), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 2), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 8), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 9)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7)) < 8)) && (1 <= floormod((threadIdx.x_1*12), 9))) && (floormod((threadIdx.x_1*12), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod((threadIdx.x_1*12), 9)) - 1)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 10)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 1), 9))) && (floormod(((threadIdx.x_1*12) + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 1), 9)) - 1)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 11)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 2), 9))) && (floormod(((threadIdx.x_1*12) + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 2), 9)) - 1)], 0f32, dtype=float32)
}
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 675)] = @tir.if_then_else(((((1 <= floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)) && (floormod(((threadIdx.x_1*6) + 27), 81) < 72)) && (1 <= floormod((threadIdx.x_1*6), 9))) && (floormod((threadIdx.x_1*6), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 225), 27)*49)) + (floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)*7)) + floormod((threadIdx.x_1*6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1176)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 6), 9))) && (floormod(((threadIdx.x_1*12) + 6), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 6), 9)) - 8)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1177)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 7), 9))) && (floormod(((threadIdx.x_1*12) + 7), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 7), 9)) - 8)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1178)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 8), 9))) && (floormod(((threadIdx.x_1*12) + 8), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 8), 9)) - 8)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1179)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7)) < 8)) && (1 <= floormod((threadIdx.x_1*12), 9))) && (floormod((threadIdx.x_1*12), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod((threadIdx.x_1*12), 9)) + 909)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1180)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 1), 9))) && (floormod(((threadIdx.x_1*12) + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 1), 9)) + 909)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1181)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 2), 9))) && (floormod(((threadIdx.x_1*12) + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 2), 9)) + 909)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1182)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 3), 9))) && (floormod(((threadIdx.x_1*12) + 3), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 394), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 3), 9)) - 8)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1183)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 4), 9))) && (floormod(((threadIdx.x_1*12) + 4), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 394), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 4), 9)) - 8)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1184)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 5), 9))) && (floormod(((threadIdx.x_1*12) + 5), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 394), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 5), 9)) - 8)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1185)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 6), 9))) && (floormod(((threadIdx.x_1*12) + 6), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 6), 9)) - 1)], 0f32, dty [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1186)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 7), 9))) && (floormod(((threadIdx.x_1*12) + 7), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 7), 9)) - 1)], 0f32, dty [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1187)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 8), 9))) && (floormod(((threadIdx.x_1*12) + 8), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 8), 9)) - 1)], 0f32, dty [...]
+ }
}
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 676)] = @tir.if_then_else(((((1 <= floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)) && (floormod(((threadIdx.x_1*6) + 28), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 1), 9))) && (floormod(((threadIdx.x_1*6) + 1), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 225), 27)*49)) + (floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)*7)) + floormod(((threadIdx.x_1*6) + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope="shared")[threadIdx.x_2] = kernel[((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 96)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 96), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 49), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 98), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 196)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 98), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 196), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 294)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 147), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 2), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 196), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 392), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 490)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 245), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 490), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 588)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 294), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 4), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 686)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 343), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 686), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 392), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 784), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 882)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 441), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 6), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 980)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 490), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 980), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1078)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 539), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1078), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1176)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 588), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1274)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 637), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1274), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1372)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 686), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1372), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1470)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 735), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 10), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 784), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1568), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1666)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 833), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1666), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1764)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 882), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 12), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1862)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 931), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1862), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1960)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 980), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1960), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2058)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1029), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 14), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2156)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1078), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2156), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2254)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1127), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2254), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2352)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1176), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 16), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2450)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1225), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2450), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2548)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1274), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2548), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2646)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1323), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 18), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2744)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1372), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2744), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2842)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1421), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2842), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2940)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1470), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 20), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ if @tir.likely((threadIdx.x_2 < 34), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 3038)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1519), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 3038), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
}
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 677)] = @tir.if_then_else(((((1 <= floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)) && (floormod(((threadIdx.x_1*6) + 29), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 2), 9))) && (floormod(((threadIdx.x_1*6) + 2), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 225), 27)*49)) + (floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)*7)) + floormod(((threadIdx.x_1*6) + 2), 9)) - 8)], 0f32, dtype=float32)
+ for (rc.outer.inner: int32, 0, 4) {
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24))]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 96)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 97)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 2)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 98)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 3)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 99)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 4)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 100)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 5)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 101)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 6)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 102)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 7)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 103)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 8)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 104)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 9)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 105)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 10)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 106)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 11)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 107)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 12)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 108)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 13)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 109)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 14)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 110)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 15)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 111)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 16)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 112)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 17)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 113)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 18)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 114)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 19)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 115)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 20)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 116)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 21)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 117)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 22)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 118)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 23)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 119)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 192)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 288)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 193)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 289)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 194)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 290)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 195)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 291)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 196)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 292)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 197)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 293)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 198)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 294)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 199)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 295)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 200)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 296)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 201)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 297)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 202)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 298)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 203)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 299)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 204)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 300)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 205)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 301)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 206)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 302)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 207)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 303)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 208)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 304)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 209)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 305)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 210)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 306)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 211)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 307)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 212)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 308)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 213)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 309)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 214)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 310)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 215)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 311)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 384)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 480)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 385)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 481)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 386)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 482)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 387)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 483)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 388)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 484)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 389)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 485)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 390)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 486)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 391)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 487)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 392)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 488)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 393)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 489)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 394)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 490)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 395)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 491)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 396)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 492)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 397)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 493)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 398)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 494)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 399)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 495)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 400)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 496)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 401)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 497)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 402)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 498)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 403)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 499)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 404)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 500)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 405)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 501)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 406)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 502)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 407)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 503)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 576)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 672)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 577)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 673)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 578)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 674)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 579)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 675)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 580)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 676)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 581)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 677)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 582)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 678)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 583)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 679)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 584)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 680)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 585)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 681)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 586)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 682)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 587)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 683)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 588)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 684)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 589)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 685)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 590)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 686)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 591)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 687)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 592)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 688)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 593)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 689)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 594)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 690)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 595)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 691)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 596)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 692)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 597)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 693)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 598)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 694)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 599)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 695)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 768)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 864)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 769)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 865)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 770)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 866)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 771)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 867)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 772)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 868)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 773)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 869)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 774)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 870)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 775)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 871)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 776)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 872)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 777)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 873)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 778)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 874)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 779)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 875)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 780)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 876)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 781)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 877)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 782)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 878)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 783)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 879)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 784)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 880)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 785)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 881)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 786)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 882)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 787)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 883)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 788)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 884)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 789)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 885)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 790)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 886)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 791)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 887)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 960)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1056)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 961)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1057)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 962)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1058)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 963)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1059)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 964)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1060)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 965)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1061)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 966)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1062)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 967)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1063)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 968)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1064)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 969)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1065)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 970)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1066)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 971)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1067)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 972)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1068)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 973)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1069)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 974)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1070)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 975)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1071)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 976)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1072)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 977)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1073)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 978)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1074)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 979)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1075)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 980)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1076)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 981)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1077)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 982)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1078)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 983)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1079)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1152)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1248)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1153)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1249)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1154)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1250)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1155)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1251)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1156)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1252)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1157)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1253)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1158)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1254)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1159)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1255)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1160)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1256)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1161)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1257)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1162)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1258)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1163)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1259)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1164)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1260)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1165)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1261)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1166)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1262)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1167)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1263)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1168)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1264)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1169)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1265)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1170)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1266)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1171)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1267)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1172)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1268)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1173)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1269)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1174)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1270)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1175)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1271)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1344)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1440)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1345)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1441)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1346)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1442)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1347)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1443)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1348)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1444)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1349)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1445)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1350)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1446)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1351)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1447)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1352)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1448)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1353)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1449)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1354)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1450)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1355)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1451)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1356)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1452)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1357)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1453)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1358)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1454)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1359)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1455)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1360)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1456)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1361)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1457)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1362)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1458)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1363)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1459)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1364)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1460)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1365)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1461)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1366)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1462)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1367)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1463)]))
}
}
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1: Buffer(kernel.shared, float32, [2304], [], scope="shared")[threadIdx.x_2] = kernel[(((blockIdx.x*73728) + cse_var_1) + threadIdx.x_2)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 7), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 112), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 14), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 80), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 21), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 28), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 35), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 128), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 42), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 96), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 49), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 56), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1008)] = kernel[((((blockIdx.x*73728) + cse_var_1) + threadIdx.x_2) + 32256)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 70), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 112), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1232)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 77), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 80), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 84), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1456)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 91), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 98), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 128), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1680)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 105), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 96), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 112), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1904)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 119), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel[((((blockIdx.x*73728) + cse_var_1) + threadIdx.x_2) + 64512)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 2128)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 133), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 112), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- if @tir.likely((threadIdx.x_2 < 64), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 140), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 80), 144))]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 324)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 405)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 486)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 333)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 414)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 495)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 576)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 342)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 423)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 585)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 351)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 432)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 513)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 594)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 360)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 522)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 603)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 369)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 450)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 531)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 612)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 459)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 540)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 621)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 325)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 487)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 568)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 334)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 415)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 496)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 577)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 424)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 505)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 586)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 352)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 433)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 514)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 595)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 361)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 523)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 604)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 370)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 451)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 532)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 613)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 460)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 541)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 622)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 326)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 407)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 488)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 569)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 335)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 416)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 497)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 578)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 344)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 425)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 506)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 587)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 353)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 515)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 596)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 362)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 524)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 605)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 452)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 533)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 614)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 461)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 542)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 623)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 333)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 414)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 495)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 576)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 342)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 423)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 585)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 351)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 432)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 513)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 594)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 360)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 522)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 603)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 369)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 450)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 531)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 612)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 459)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 540)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 621)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 387)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 468)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 549)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 334)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 415)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 496)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 577)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 424)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 505)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 586)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 352)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 433)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 514)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 595)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 361)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 523)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 604)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 370)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 451)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 532)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 613)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 460)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 541)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 622)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 388)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 550)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 631)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 335)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 416)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 497)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 578)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 344)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 425)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 506)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 587)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 353)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 515)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 596)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 362)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 524)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 605)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 452)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 533)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 614)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 461)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 542)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 623)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 389)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 470)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 551)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 632)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 342)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 423)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 585)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 351)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 432)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 513)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 594)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 360)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 522)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 603)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 369)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 450)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 531)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 612)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 459)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 540)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 621)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 387)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 468)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 549)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 396)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 477)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 558)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 639)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 424)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 505)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 586)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 352)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 433)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 514)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 595)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 361)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 523)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 604)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 370)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 451)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 532)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 613)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 460)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 541)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 622)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 388)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 550)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 631)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 397)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 478)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 559)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 640)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 344)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 425)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 506)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 587)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 353)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 515)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 596)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 362)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 524)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 605)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 452)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 533)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 614)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 461)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 542)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 623)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 389)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 470)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 551)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 632)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 398)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 479)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 560)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 641)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 648)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 729)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 810)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 891)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 972)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1053)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1215)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 657)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 738)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 900)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 981)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1062)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1143)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 666)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 747)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 828)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 909)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 990)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1152)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1233)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 675)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 837)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 918)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 999)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1080)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1242)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 684)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 765)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 846)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 927)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1089)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1170)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1251)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 774)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 855)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 936)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1017)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1098)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1179)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 702)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 783)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 864)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1026)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1107)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1188)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1269)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 649)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 730)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 811)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 892)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 973)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1054)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 658)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 739)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 820)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 901)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 982)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1063)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 667)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 748)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 829)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 910)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 991)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1072)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 676)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 757)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 838)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 919)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1000)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1081)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 685)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 766)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 847)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 928)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1009)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1090)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 694)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 775)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 856)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 937)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1018)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1099)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 703)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 784)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 865)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 946)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1027)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 650)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 731)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 812)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 893)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 974)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1055)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 659)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 740)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 821)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 902)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 983)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1064)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 668)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 749)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 830)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 911)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 992)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1073)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 677)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 758)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 839)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 920)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1001)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1082)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 686)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 767)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 848)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 929)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1010)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1091)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 695)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 776)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 857)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 938)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1019)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 704)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 785)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 866)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 947)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1028)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 657)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 738)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 900)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 981)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1062)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1143)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 666)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 747)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 828)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 909)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 990)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1152)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1233)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 675)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 837)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 918)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 999)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1080)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1242)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 684)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 765)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 846)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 927)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1089)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1170)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1251)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 774)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 855)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 936)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1017)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1098)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1179)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 702)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 783)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 864)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1026)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1107)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1188)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1269)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 711)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 792)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 873)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 954)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1035)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1116)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1278)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 658)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 739)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 820)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 901)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 982)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1063)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 667)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 748)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 829)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 910)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 991)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1072)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 676)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 757)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 838)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 919)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1000)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1081)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 685)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 766)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 847)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 928)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1009)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1090)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 694)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 775)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 856)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 937)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1018)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1099)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 703)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 784)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 865)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 946)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1027)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 712)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 793)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 874)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 955)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1036)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 659)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 740)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 821)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 902)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 983)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1064)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 668)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 749)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 830)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 911)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 992)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1073)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 677)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 758)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 839)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 920)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1001)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1082)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 686)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 767)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 848)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 929)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1010)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1091)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 695)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 776)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 857)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 938)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1019)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 704)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 785)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 866)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 947)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1028)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 713)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 794)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 875)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 956)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1037)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 666)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 747)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 828)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 909)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 990)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1152)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1233)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 675)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 837)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 918)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 999)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1080)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1242)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 684)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 765)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 846)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 927)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1089)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1170)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1251)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 774)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 855)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 936)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1017)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1098)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1179)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 702)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 783)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 864)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1026)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1107)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1188)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1269)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 711)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 792)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 873)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 954)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1035)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1116)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1278)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 720)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 801)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 963)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1044)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1125)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1206)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 667)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 748)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 829)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 910)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 991)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1072)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 676)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 757)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 838)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 919)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1000)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1081)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 685)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 766)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 847)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 928)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1009)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1090)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 694)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 775)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 856)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 937)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1018)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1099)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 703)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 784)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 865)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 946)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1027)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 712)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 793)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 874)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 955)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1036)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 721)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 802)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 883)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 964)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1045)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 668)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 749)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 830)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 911)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 992)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1073)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 677)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 758)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 839)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 920)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1001)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1082)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 686)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 767)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 848)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 929)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1010)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1091)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 695)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 776)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 857)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 938)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1019)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 704)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 785)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 866)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 947)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1028)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 713)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 794)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 875)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 956)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1037)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 722)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 803)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 884)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 965)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1046)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
}
}
- for (i2.inner: int32, 0, 7) {
- compute[((((blockIdx.x*784) + (floordiv(threadIdx.x, 7)*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[i2.inner] + bias[((blockIdx.x*16) + floordiv(threadIdx.x, 7))]), 0f32)
+ for (i1.inner: int32, 0, 16) {
+ compute[((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*784)) + (i1.inner*49)) + floormod(threadIdx.x, 49))] = max((conv2d_nchw_1[i1.inner] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*16)) + i1.inner)]), 0f32)
}
}
}
@@ -1373,7 +810,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.155 ms
+ Execution time of this operator: 0.274 ms
@@ -1417,33 +854,33 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
- conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
- conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
- conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=16)
+ conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
+ conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=8)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=2)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
- conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=7)
- conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
+ conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
+ conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=8)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
- conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
- conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
- conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
+ conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
+ conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
+ conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
- compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=16)
+ compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=16)
+ compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=2)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
- compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=7)
- compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
+ compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+ compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
@@ -1466,12 +903,12 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
+ kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=6)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=12)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 1024)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
@@ -1491,10 +928,10 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
#define int64_t long long
#define uint64_t unsigned long long
#endif
- extern "C" __global__ void __launch_bounds__(112) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[7];
- __shared__ float pad_temp_shared[1296];
- __shared__ float kernel_shared[2304];
+ extern "C" __global__ void __launch_bounds__(98) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[16];
+ __shared__ float pad_temp_shared[2016];
+ __shared__ float kernel_shared[3072];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
@@ -1502,1067 +939,491 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw[4] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 32; ++rc_outer_outer) {
- __syncthreads();
- pad_temp_shared[(((int)threadIdx.x) * 6)] = (((((3 <= ((((int)threadIdx.x) * 2) % 27)) && (((((int)threadIdx.x) * 6) % 81) < 72)) && (1 <= ((((int)threadIdx.x) * 6) % 9))) && (((((int)threadIdx.x) * 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 2) / 27) * 49)) + ((((((int)threadIdx.x) * 2) % 27) / 3) * 7)) + ((((int)threadIdx.x) * 6) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 1)] = (((((3 <= ((((int)threadIdx.x) * 2) % 27)) && ((((((int)threadIdx.x) * 6) + 1) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 1) % 9))) && ((((((int)threadIdx.x) * 6) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 2) / 27) * 49)) + ((((((int)threadIdx.x) * 2) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 1) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 2)] = (((((3 <= ((((int)threadIdx.x) * 2) % 27)) && ((((((int)threadIdx.x) * 6) + 2) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 2) % 9))) && ((((((int)threadIdx.x) * 6) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 2) / 27) * 49)) + ((((((int)threadIdx.x) * 2) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 2) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 3)] = (((((3 <= (((((int)threadIdx.x) * 2) + 1) % 27)) && ((((((int)threadIdx.x) * 6) + 3) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 3) % 9))) && ((((((int)threadIdx.x) * 6) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 1) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 3) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 4)] = (((((3 <= (((((int)threadIdx.x) * 2) + 1) % 27)) && ((((((int)threadIdx.x) * 6) + 4) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 4) % 9))) && ((((((int)threadIdx.x) * 6) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 1) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 4) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 5)] = (((((3 <= (((((int)threadIdx.x) * 2) + 1) % 27)) && ((((((int)threadIdx.x) * 6) + 5) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 5) % 9))) && ((((((int)threadIdx.x) * 6) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 1) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 5) % 9)) - 8)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 672)] = (((((3 <= (((((int)threadIdx.x) * 2) + 8) % 27)) && ((((((int)threadIdx.x) * 6) + 24) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 6) % 9))) && ((((((int)threadIdx.x) * 6) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 224) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 8) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 6) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 673)] = (((((3 <= (((((int)threadIdx.x) * 2) + 8) % 27)) && ((((((int)threadIdx.x) * 6) + 25) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 7) % 9))) && ((((((int)threadIdx.x) * 6) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 224) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 8) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 7) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 674)] = (((((3 <= (((((int)threadIdx.x) * 2) + 8) % 27)) && ((((((int)threadIdx.x) * 6) + 26) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 8) % 9))) && ((((((int)threadIdx.x) * 6) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 224) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 8) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 8) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 675)] = (((((1 <= ((((((int)threadIdx.x) * 2) / 3) + 3) % 9)) && ((((((int)threadIdx.x) * 6) + 27) % 81) < 72)) && (1 <= ((((int)threadIdx.x) * 6) % 9))) && (((((int)threadIdx.x) * 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 225) / 27) * 49)) + (((((((int)threadIdx.x) * 2) / 3) + 3) % 9) * 7)) + ((((int)threadIdx.x) * 6) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 676)] = (((((1 <= ((((((int)threadIdx.x) * 2) / 3) + 3) % 9)) && ((((((int)threadIdx.x) * 6) + 28) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 1) % 9))) && ((((((int)threadIdx.x) * 6) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 225) / 27) * 49)) + (((((((int)threadIdx.x) * 2) / 3) + 3) % 9) * 7)) + (((((int)threadIdx.x) * 6) + 1) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 677)] = (((((1 <= ((((((int)threadIdx.x) * 2) / 3) + 3) % 9)) && ((((((int)threadIdx.x) * 6) + 29) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 2) % 9))) && ((((((int)threadIdx.x) * 6) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 225) / 27) * 49)) + (((((((int)threadIdx.x) * 2) / 3) + 3) % 9) * 7)) + (((((int)threadIdx.x) * 6) + 2) % 9)) - 8)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 73728) + (rc_outer_outer * 144)) + ((int)threadIdx.x))];
- kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 112) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 112) % 144))];
- kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 224) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 80) % 144))];
- kernel_shared[(((int)threadIdx.x) + 336)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 336) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 48) % 144))];
- kernel_shared[(((int)threadIdx.x) + 448)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 448) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 16))];
- kernel_shared[(((int)threadIdx.x) + 560)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 560) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 128) % 144))];
- kernel_shared[(((int)threadIdx.x) + 672)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 672) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 96) % 144))];
- kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 784) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 64) % 144))];
- kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 896) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 32))];
- kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 32256)];
- kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1120) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 112) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1232) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 80) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1344) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 48) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1456) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 16))];
- kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1568) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 128) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1680) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 96) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1792) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 64) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1904) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 32))];
- kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 64512)];
- kernel_shared[(((int)threadIdx.x) + 2128)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2128) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 112) % 144))];
- if (((int)threadIdx.x) < 64) {
- kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2240) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 80))];
+ conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
+ for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[(((int)threadIdx.x) * 12)] = (((((1 <= ((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) * 12) % 9))) && (((((int)threadIdx.x) * 12) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) * 12) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1)] = (((((1 <= ((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 1) % 9))) && ((((((int)threadIdx.x) * 12) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 2)] = (((((1 <= ((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 2) % 9))) && ((((((int)threadIdx.x) * 12) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 3)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 3) % 9))) && ((((((int)threadIdx.x) * 12) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 4)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 4) % 9))) && ((((((int)threadIdx.x) * 12) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 5)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 5) % 9))) && ((((((int)threadIdx.x) * 12) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 6)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 6) % 9))) && ((((((int)threadIdx.x) * 12) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 7)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 7) % 9))) && ((((((int)threadIdx.x) * 12) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 8)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 8) % 9))) && ((((((int)threadIdx.x) * 12) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 9)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 12) % 9))) && (((((int)threadIdx.x) * 12) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) * 12) % 9)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 10)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 1) % 9))) && ((((((int)threadIdx.x) * 12) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 1) % 9)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 11)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 2) % 9))) && ((((((int)threadIdx.x) * 12) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 2) % 9)) - 1)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1176)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 6) % 9))) && ((((((int)threadIdx.x) * 12) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 6) % 9)) - 8)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1177)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 7) % 9))) && ((((((int)threadIdx.x) * 12) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 7) % 9)) - 8)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1178)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 8) % 9))) && ((((((int)threadIdx.x) * 12) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 8) % 9)) - 8)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1179)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 12) % 9))) && (((((int)threadIdx.x) * 12) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) * 12) % 9)) + 909)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1180)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 1) % 9))) && ((((((int)threadIdx.x) * 12) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 1) % 9)) + 909)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1181)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 2) % 9))) && ((((((int)threadIdx.x) * 12) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 2) % 9)) + 909)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1182)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 3) % 9))) && ((((((int)threadIdx.x) * 12) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 394) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 3) % 9)) - 8)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1183)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 4) % 9))) && ((((((int)threadIdx.x) * 12) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 394) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 4) % 9)) - 8)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1184)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 5) % 9))) && ((((((int)threadIdx.x) * 12) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 394) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 5) % 9)) - 8)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1185)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 6) % 9))) && ((((((int)threadIdx.x) * 12) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 6) % 9)) - 1)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1186)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 7) % 9))) && ((((((int)threadIdx.x) * 12) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 7) % 9)) - 1)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1187)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 8) % 9))) && ((((((int)threadIdx.x) * 12) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 8) % 9)) - 1)] : 0.000000e+00f);
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 98)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 98) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 2) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 196) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 4) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 294)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 294) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 2) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 8) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 490)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 490) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 10) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 588)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 588) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 4) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 686)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 686) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 14) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 16) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 882)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 882) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 6) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 980)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 980) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 20) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1078)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1078) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 22) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1176) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 8) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1274)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1274) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 26) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1372) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 28) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1470)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1470) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 10) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1568) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 32) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1666)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1666) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 34) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1764)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1764) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 12) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1862)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1862) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 38) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1960) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 40) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2058)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2058) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 14) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2156)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2156) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 44) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2254)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2254) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 46) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2352) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2450)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2450) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 50) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2548)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2548) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 52) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2646)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2646) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 18) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2744) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2842)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2842) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 58) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2940)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2940) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 20) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ if (((int)threadIdx.x) < 34) {
+ kernel_shared[(((int)threadIdx.x) + 3038)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3038) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 62) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ }
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 4; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24))]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 96)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 97)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 2)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 98)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 3)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 99)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 4)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 100)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 5)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 101)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 6)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 102)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 7)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 103)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 8)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 104)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 9)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 105)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 10)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 106)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 11)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 107)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 12)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 108)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 13)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 109)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 14)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 110)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 15)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 111)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 16)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 112)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 17)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 113)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 18)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 114)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 19)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 115)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 20)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 116)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 21)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 117)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 22)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 118)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 23)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 119)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 192)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 288)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 193)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 289)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 194)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 290)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 195)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 291)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 196)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 292)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 197)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 293)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 198)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 294)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 199)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 295)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 200)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 296)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 201)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 297)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 202)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 298)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 203)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 299)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 204)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 300)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 205)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 301)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 206)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 302)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 207)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 303)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 208)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 304)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 209)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 305)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 210)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 306)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 211)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 307)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 212)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 308)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 213)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 309)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 214)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 310)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 215)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 311)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 384)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 480)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 385)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 481)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 386)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 482)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 387)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 483)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 388)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 484)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 389)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 485)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 390)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 486)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 391)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 487)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 392)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 488)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 393)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 489)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 394)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 490)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 395)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 491)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 396)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 492)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 397)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 493)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 398)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 494)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 399)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 495)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 400)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 496)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 401)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 497)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 402)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 498)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 403)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 499)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 404)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 500)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 405)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 501)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 406)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 502)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 407)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 503)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 576)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 672)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 577)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 673)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 578)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 674)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 579)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 675)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 580)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 676)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 581)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 677)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 582)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 678)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 583)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 679)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 584)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 680)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 585)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 681)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 586)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 682)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 587)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 683)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 588)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 684)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 589)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 685)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 590)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 686)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 591)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 687)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 592)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 688)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 593)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 689)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 594)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 690)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 595)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 691)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 596)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 692)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 597)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 693)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 598)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 694)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 599)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 695)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 768)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 864)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 769)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 865)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 770)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 866)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 771)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 867)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 772)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 868)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 773)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 869)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 774)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 870)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 775)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 871)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 776)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 872)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 777)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 873)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 778)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 874)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 779)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 875)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 780)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 876)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 781)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 877)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 782)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 878)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 783)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 879)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 784)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 880)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 785)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 881)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 786)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 882)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 787)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 883)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 788)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 884)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 789)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 885)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 790)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 886)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 791)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 887)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 960)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1056)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 961)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1057)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 962)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1058)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 963)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1059)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 964)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1060)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 965)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1061)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 966)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1062)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 967)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1063)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 968)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1064)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 969)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1065)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 970)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1066)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 971)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1067)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 972)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1068)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 973)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1069)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 974)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1070)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 975)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1071)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 976)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1072)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 977)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1073)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 978)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1074)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 979)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1075)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 980)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1076)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 981)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1077)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 982)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1078)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 983)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1079)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1152)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1248)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1153)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1249)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1154)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1250)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1155)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1251)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1156)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1252)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1157)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1253)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1158)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1254)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1159)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1255)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1160)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1256)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1161)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1257)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1162)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1258)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1163)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1259)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1164)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1260)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1165)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1261)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1166)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1262)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1167)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1263)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1168)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1264)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1169)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1265)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1170)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1266)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1171)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1267)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1172)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1268)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1173)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1269)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1174)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1270)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1175)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1271)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1344)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1440)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1345)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1441)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1346)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1442)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1347)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1443)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1348)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1444)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1349)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1445)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1350)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1446)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1351)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1447)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1352)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1448)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1353)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1449)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1354)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1450)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1355)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1451)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1356)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1452)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1357)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1453)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1358)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1454)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1359)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1455)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1360)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1456)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1361)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1457)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1362)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1458)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1363)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1459)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1364)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1460)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1365)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1461)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1366)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1462)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1367)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1463)]));
+ }
}
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 324)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 405)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 486)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 333)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 414)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 495)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 576)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 342)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 423)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 585)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 351)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 432)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 513)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 594)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 360)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 522)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 603)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 369)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 450)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 531)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 612)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 459)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 540)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 621)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 325)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 487)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 334)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 415)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 496)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 577)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 424)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 586)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 352)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 433)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 514)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 595)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 361)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 523)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 604)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 370)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 451)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 532)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 613)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 460)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 541)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 622)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 326)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 407)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 488)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 335)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 416)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 497)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 578)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 344)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 425)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 587)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 353)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 434)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 515)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 596)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 362)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 524)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 605)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 371)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 452)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 533)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 614)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 461)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 542)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 623)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 333)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 414)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 495)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 576)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 342)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 423)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 585)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 351)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 432)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 513)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 594)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 360)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 522)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 603)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 369)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 450)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 531)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 612)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 459)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 540)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 621)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 387)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 468)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 549)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 334)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 415)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 496)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 577)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 424)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 586)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 352)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 433)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 514)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 595)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 361)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 523)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 604)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 370)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 451)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 532)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 613)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 460)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 541)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 622)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 388)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 550)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 335)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 416)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 497)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 578)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 344)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 425)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 587)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 353)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 434)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 515)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 596)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 362)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 524)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 605)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 371)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 452)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 533)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 614)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 461)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 542)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 623)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 389)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 470)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 551)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 342)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 423)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 585)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 351)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 432)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 513)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 594)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 360)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 522)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 603)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 369)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 450)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 531)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 612)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 459)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 540)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 621)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 387)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 468)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 549)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 396)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 477)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 558)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 639)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 424)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 586)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 352)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 433)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 514)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 595)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 361)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 523)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 604)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 370)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 451)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 532)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 613)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 460)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 541)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 622)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 388)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 550)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 397)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 478)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 559)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 640)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 344)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 425)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 587)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 353)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 434)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 515)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 596)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 362)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 524)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 605)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 371)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 452)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 533)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 614)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 461)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 542)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 623)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 389)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 470)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 551)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 398)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 479)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 560)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 641)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 648)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 729)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 810)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 891)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 972)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1053)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1215)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 657)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 738)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 819)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 900)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 981)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1062)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1143)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 666)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 747)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 828)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 909)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 990)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1152)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 675)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 756)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 837)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 918)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 999)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1080)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1242)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 684)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 765)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 846)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 927)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1089)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1251)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 693)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 774)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 855)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 936)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1017)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1098)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1179)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 702)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 783)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 864)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 945)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1026)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1188)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1269)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 649)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 730)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 811)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 892)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 973)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1054)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 658)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 739)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 820)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 901)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 982)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1063)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 667)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 748)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 829)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 910)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 991)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1072)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 676)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 757)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 838)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 919)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1000)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1081)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 685)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 766)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 847)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 928)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1009)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1090)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 694)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 775)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 856)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 937)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1018)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1099)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 703)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 784)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 865)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 946)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1027)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 650)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 731)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 812)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 893)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 974)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1055)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 659)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 740)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 821)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 902)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 983)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1064)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 668)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 749)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 830)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 911)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 992)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1073)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 677)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 758)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 839)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 920)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1001)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1082)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 686)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 767)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 848)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 929)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1010)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1091)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 695)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 776)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 857)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 938)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1019)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 704)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 785)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 866)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 947)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1028)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 657)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 738)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 819)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 900)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 981)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1062)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1143)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 666)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 747)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 828)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 909)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 990)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1152)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 675)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 756)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 837)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 918)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 999)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1080)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1242)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 684)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 765)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 846)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 927)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1089)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1251)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 693)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 774)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 855)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 936)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1017)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1098)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1179)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 702)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 783)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 864)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 945)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1026)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1188)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1269)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 711)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 792)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 873)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 954)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1035)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1116)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1278)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 658)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 739)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 820)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 901)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 982)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1063)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 667)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 748)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 829)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 910)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 991)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1072)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 676)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 757)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 838)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 919)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1000)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1081)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 685)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 766)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 847)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 928)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1009)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1090)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 694)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 775)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 856)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 937)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1018)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1099)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 703)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 784)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 865)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 946)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1027)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 712)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 793)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 874)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 955)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1036)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 659)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 740)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 821)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 902)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 983)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1064)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 668)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 749)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 830)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 911)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 992)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1073)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 677)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 758)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 839)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 920)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1001)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1082)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 686)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 767)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 848)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 929)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1010)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1091)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 695)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 776)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 857)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 938)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1019)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 704)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 785)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 866)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 947)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1028)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 713)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 794)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 875)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 956)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1037)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 666)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 747)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 828)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 909)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 990)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1152)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 675)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 756)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 837)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 918)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 999)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1080)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1242)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 684)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 765)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 846)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 927)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1089)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1251)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 693)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 774)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 855)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 936)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1017)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1098)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1179)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 702)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 783)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 864)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 945)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1026)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1188)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1269)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 711)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 792)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 873)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 954)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1035)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1116)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1278)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 720)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 801)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 882)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 963)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1044)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1125)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1206)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 667)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 748)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 829)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 910)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 991)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1072)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 676)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 757)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 838)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 919)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1000)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1081)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 685)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 766)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 847)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 928)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1009)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1090)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 694)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 775)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 856)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 937)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1018)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1099)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 703)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 784)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 865)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 946)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1027)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 712)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 793)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 874)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 955)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1036)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 721)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 802)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 883)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 964)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1045)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 668)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 749)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 830)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 911)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 992)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1073)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 677)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 758)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 839)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 920)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1001)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1082)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 686)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 767)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 848)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 929)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1010)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1091)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 695)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 776)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 857)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 938)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1019)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 704)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 785)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 866)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 947)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1028)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 713)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 794)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 875)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 956)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1037)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 722)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 803)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 884)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 965)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1046)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
}
- for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
- compute[((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 7) * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[i2_inner] + bias[((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 16; ++i1_inner) {
+ compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 784)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 16)) + i1_inner)]), 0.000000e+00f);
}
}
@@ -2621,7 +1482,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 26.864 seconds)
+ **Total running time of the script:** ( 2 minutes 25.965 seconds)
.. _sphx_glr_download_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py:
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
index 5a79ed0..c60a323 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
@@ -614,7 +614,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 9.8594 9.8517 9.9018 9.8247 0.0320
+ 9.9032 9.9073 9.9324 9.8701 0.0256
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
index ab7ea3d..c87ed03 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
@@ -633,7 +633,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 760.1292 760.3889 760.6901 759.3086 0.5931
+ 748.3389 746.1669 752.9568 745.8929 3.2673
@@ -658,7 +658,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 20.793 seconds)
+ **Total running time of the script:** ( 1 minutes 18.612 seconds)
.. _sphx_glr_download_how_to_tune_with_autoscheduler_tune_network_x86.py:
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
index 9b9ad9f..574ecda 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
@@ -355,36 +355,33 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
Lowered TIR:
@main = primfn(placeholder_5: handle, placeholder_6: handle, placeholder_7: handle, placeholder_8: handle, placeholder_9: handle, compute_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {placeholder_4: Buffer(placeholder_10: Pointer(float32), float32, [32768], []),
- placeholder_2: Buffer(placeholder_11: Pointer(float32), float32, [78656], []),
- placeholder_3: Buffer(placeholder_12: Pointer(int32), int32, [4916], []),
+ buffers = {placeholder_4: Buffer(placeholder_10: Pointer(int32), int32, [33], []),
+ placeholder_2: Buffer(placeholder_11: Pointer(int32), int32, [4916], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], []),
- placeholder: Buffer(placeholder_13: Pointer(float32), float32, [65536], []),
- placeholder_1: Buffer(placeholder_14: Pointer(int32), int32, [33], [])}
- buffer_map = {placeholder_9: placeholder, placeholder_8: placeholder_1, compute_1: compute, placeholder_6: placeholder_2, placeholder_7: placeholder_3, placeholder_5: placeholder_4} {
- allocate(compute_3: Pointer(global float32), float32, [4096]), storage_scope = global;
- for (i1.outer: int32, 0, 16) {
- for (i.outer.inner: int32, 0, 16) {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 8) {
- for (j.init: int32, 0, 16) {
- compute_4: Buffer(compute_3, float32, [4096], [])[((((i.outer.inner*256) + (i.inner.init*32)) + (nb_j.inner*16)) + j.init)] = 0f32
- }
+ placeholder: Buffer(placeholder_12: Pointer(float32), float32, [32768], []),
+ placeholder_3: Buffer(placeholder_13: Pointer(float32), float32, [65536], []),
+ placeholder_1: Buffer(placeholder_14: Pointer(float32), float32, [78656], [])}
+ buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, compute_1: compute, placeholder_7: placeholder_2, placeholder_9: placeholder_3, placeholder_8: placeholder_4} {
+ for (i0.outer.i1.outer.fused: int32, 0, 256) "parallel" {
+ allocate(compute_3: Pointer(global float32), float32, [256]), storage_scope = global {
+ for (i.inner.init: int32, 0, 16) {
+ for (j.init: int32, 0, 16) {
+ compute_4: Buffer(compute_3, float32, [256], [])[((i.inner.init*16) + j.init)] = 0f32
}
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((i1.outer*2) + nb_j.inner) in (placeholder_1[(cse_var_1 + 1)] - placeholder_1[cse_var_1])) {
- for (i.inner: int32, 0, 8) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((i1.outer*2) + nb_j.inner)
- let cse_var_2: int32 = ((((i.outer.inner*256) + (i.inner*32)) + (nb_j.inner*16)) + j)
- compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_2[(((placeholder_1[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder_4[(((i.outer.inner*2048) + (i.inner*256)) + placeholder_3[(placeholder_1[cse_var_3] + elem_idx)])], 0f32)))
- }
+ }
+ for (elem_idx: int32, 0, let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_4[(cse_var_1 + 1)] - placeholder_4[cse_var_1])) {
+ for (i.inner: int32, 0, 16) {
+ for (j: int32, 0, 16) {
+ let cse_var_3: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ let cse_var_2: int32 = ((i.inner*16) + j)
+ compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_1[(((placeholder_4[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + (i.inner*256)) + placeholder_2[(placeholder_4[cse_var_3] + elem_idx)])], 0f32)))
}
}
}
- }
- for (i0.inner: int32, 0, 128) {
- let cse_var_4: int32 = ((i0.inner*512) + (i1.outer*32))
- compute[ramp(cse_var_4, 1, 32)] = max((compute_4[ramp((i0.inner*32), 1, 32)] + placeholder[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 16) {
+ let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*8192) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
+ compute[ramp(cse_var_4, 1, 16)] = max((compute_4[ramp((i0.inner*16), 1, 16)] + placeholder_3[ramp(cse_var_4, 1, 16)]), broadcast(0f32, 16))
+ }
}
}
}
@@ -437,7 +434,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 1.411 ms
+ Execution time of this operator: 1.538 ms
diff --git a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
index 195dd37..dfe3fa1 100644
--- a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:44.397** total execution time for **how_to_tune_with_autotvm** files:
+**00:43.813** total execution time for **how_to_tune_with_autotvm** files:
-- **00:43.520**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)
-- **00:00.230**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)
-- **00:00.217**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)
-- **00:00.216**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)
-- **00:00.215**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``)
+- **00:43.071**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)
+- **00:00.196**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)
+- **00:00.184**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``)
+- **00:00.183**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)
+- **00:00.180**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)
diff --git a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
index bc30aeb..20bc8ef 100644
--- a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
@@ -859,8 +859,8 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2885496
- No: 6 GFLOPS: 100.79/100.79 result: MeasureResult(costs=(0.0022969085208333333,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5907156467437744, timestamp=1647466290.0848103) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
- No: 7 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 6 GFLOPS: 94.20/94.20 result: MeasureResult(costs=(0.0024575866875,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5868585109710693, timestamp=1647477956.84365) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
+ No: 7 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -983,7 +983,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6225319
- No: 8 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 8 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1106,7 +1106,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,943546
- No: 9 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 9 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1229,7 +1229,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2868708
- No: 10 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 10 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
res = future.result()
File "/usr/lib/python3.6/concurrent/futures/_base.py", line 432, in result
@@ -1247,7 +1247,7 @@ for this template
TimeoutError
[('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4691833
- No: 11 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 11 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1370,7 +1370,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1042124
- No: 12 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 12 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1493,7 +1493,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10013405
- No: 13 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 13 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1616,7 +1616,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6732082
- No: 14 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 14 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1739,7 +1739,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7536735
- No: 15 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 15 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1862,7 +1862,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,482121
- No: 16 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 16 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1985,7 +1985,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2824525
- No: 17 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 17 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2108,7 +2108,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4559286
- No: 18 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 18 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2231,7 +2231,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9677544
- No: 19 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+ No: 19 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 721, in __call__
yield remote, remote.load_module(os.path.split(build_result.filename)[1])
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 685, in run_through_rpc
@@ -2327,7 +2327,7 @@ for this template
15: _PyEval_EvalFrameDefault
14: 0x000000000050a7f2
13: _PyObject_FastCallKeywords
- 12: 0x00007f1aafd34c12
+ 12: 0x00007f5502403c12
11: _ctypes_callproc
10: ffi_call
9: ffi_call_unix64
@@ -2397,7 +2397,7 @@ for this template
24: 0x000000000050a6bc
23: 0x0000000000509988
22: _PyEval_EvalFrameDe [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6390073
- No: 20 GFLOPS: 144.43/144.43 result: MeasureResult(costs=(0.00160282077,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4138598442077637, timestamp=1647466315.689689) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
+ No: 20 GFLOPS: 144.18/144.18 result: MeasureResult(costs=(0.0016056626900000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3832473754882812, timestamp=1647477981.520465) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
@@ -2450,7 +2450,7 @@ and measure running time.
Best config:
[('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
- Time cost of this operator: 0.001982
+ Time cost of this operator: 0.001978
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
index 9bc6ce0..840c638 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
@@ -292,10 +292,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 313.0 98.753 (1, 2, 10, 10, 3) 2 1
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.0 0.947 (1, 6, 10, 10) 1 1
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.953 0.301 (1, 1, 10, 10, 3) 1 1
- Total_time - 316.953 - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 310.7 98.704 (1, 2, 10, 10, 3) 2 1
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.144 0.999 (1, 6, 10, 10) 1 1
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.934 0.297 (1, 1, 10, 10, 3) 1 1
+ Total_time - 314.778 - - - -
@@ -357,10 +357,10 @@ Timing the tuned program
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 92.6 97.204 (1, 6, 10, 10, 1) 2 1
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.741 1.828 (1, 6, 10, 10) 1 1
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.922 0.968 (1, 1, 10, 10, 3) 1 1
- Total_time - 95.263 - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 80.35 96.752 (1, 6, 10, 10, 1) 2 1
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.77 2.132 (1, 6, 10, 10) 1 1
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.927 1.116 (1, 1, 10, 10, 3) 1 1
+ Total_time - 83.047 - - - -
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_tflite.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_tflite.rst.txt
index 631465c..ab94888 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_tflite.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_tflite.rst.txt
@@ -365,7 +365,7 @@ Now, compile the model for the target:
#ifdef __cplusplus
extern "C"
#endif
- TVM_DLL int32_t tvmgen_default_fused_nn_dense_add(void* args, int32_t* arg_type_ids, int32_t num_args, void* out_ret_value, int32_t* out_ret_tcode, void* resource_handle) {
+ TVM_DLL int32_t tvmgen_default_fused_reshape_1(void* args, int32_t* arg_type_ids, int32_t num_args, void* out_ret_value, int32_t* out_ret_tcode, void* resource_handle) {
void* arg_placeholder = (((TVMValue*)args)[0].v_handle);
- .
- ./codegen
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_tvmc.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_tvmc.rst.txt
index 144b9c3..883426a 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_tvmc.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_tvmc.rst.txt
@@ -44,14 +44,15 @@ Using TVMC Micro
.. code-block:: bash
- tvmc --help
+ tvmc --help
To compile a model for microtvm we use ``tvmc compile`` subcommand. The output of this command
is used in next steps with ``tvmc micro`` subcommands. You can check the availability of TVMC Micro using:
.. code-block:: bash
- tvmc micro --help
+ tvmc micro --help
+
The main tasks that you can perform using ``tvmc micro`` are ``create``, ``build`` and ``flash``.
To read about specific options under a givern subcommand, use
@@ -68,7 +69,7 @@ Obtain a Tiny Model
.. code-block:: bash
- wget https://github.com/tensorflow/tflite-micro/raw/main/tensorflow/lite/micro/examples/magic_wand/magic_wand.tflite
+ wget https://github.com/tensorflow/tflite-micro/raw/main/tensorflow/lite/micro/examples/magic_wand/magic_wand.tflite
Compiling a TFLite model to a Model Library Format
@@ -82,16 +83,16 @@ Compiling a TFLite model to a Model Library Format
.. code-block:: bash
- tvmc compile magic_wand.tflite \
- --target='c -keys=cpu -link-params=0 -model=host' \
- --runtime=crt \
- --runtime-crt-system-lib 1 \
- --executor='graph' \
- --executor-graph-link-params 0 \
- --output model.tar \
- --output-format mlf \
- --pass-config tir.disable_vectorize=1 \
- --disabled-pass=AlterOpLayout
+ tvmc compile magic_wand.tflite \
+ --target='c -keys=cpu -link-params=0 -model=host' \
+ --runtime=crt \
+ --runtime-crt-system-lib 1 \
+ --executor='graph' \
+ --executor-graph-link-params 0 \
+ --output model.tar \
+ --output-format mlf \
+ --pass-config tir.disable_vectorize=1 \
+ --disabled-pass=AlterOpLayout
This will generate a ``model.tar`` file which contains TVM compiler output files. To run this command for
a different Zephyr device, you need to update ``target``. For instance, for ``nrf5340dk_nrf5340_cpuapp`` board
@@ -107,11 +108,11 @@ Create a Zephyr Project Using Model Library Format
.. code-block:: bash
- tvmc micro create \
- project \
- model.tar \
- zephyr \
- --project-option project_type=host_driven zephyr_board=qemu_x86
+ tvmc micro create \
+ project \
+ model.tar \
+ zephyr \
+ --project-option project_type=host_driven zephyr_board=qemu_x86
This will generate a ``Host-Driven`` Zephyr project for ``qemu_x86`` Zephyr board. In Host-Driven template project,
the Graph Executor will run on host and perform the model execution on Zephyr device by issuing commands to the
@@ -132,10 +133,10 @@ Build and Flash Zephyr Project Using TVMC Micro
.. code-block:: bash
- tvmc micro build \
- project \
- zephyr \
- --project-option zephyr_board=qemu_x86
+ tvmc micro build \
+ project \
+ zephyr \
+ --project-option zephyr_board=qemu_x86
This will build the project in ``project`` directory and generates binary files under ``project/build``. To build
Zephyr project for a different Zephyr board, change ``zephyr_board`` project option.
@@ -145,10 +146,10 @@ Build and Flash Zephyr Project Using TVMC Micro
.. code-block:: bash
- tvmc micro flash \
- project \
- zephyr \
- --project-option zephyr_board=qemu_x86
+ tvmc micro flash \
+ project \
+ zephyr \
+ --project-option zephyr_board=qemu_x86
Run Tiny Model on Micro Target
@@ -162,12 +163,13 @@ Run Tiny Model on Micro Target
.. code-block:: bash
- tvmc run \
- --device micro \
- project \
- --project-option zephyr_board=qemu_x86 \
- --fill-mode ones
- --print-top 4
+ tvmc run \
+ --device micro \
+ project \
+ --project-option zephyr_board=qemu_x86 \
+ --fill-mode ones \
+ --print-top 4
+
# Output:
#
# INFO:__main__:b'[100%] [QEMU] CPU: qemu32,+nx,+pae\n'
diff --git a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
index 3f75eea..12ee854 100644
--- a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:48.468** total execution time for **how_to_work_with_microtvm** files:
+**00:46.529** total execution time for **how_to_work_with_microtvm** files:
-- **00:44.158**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)
-- **00:03.716**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)
-- **00:00.202**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``)
-- **00:00.196**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)
-- **00:00.195**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tvmc.py` (``micro_tvmc.py``)
+- **00:42.449**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)
+- **00:03.571**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)
+- **00:00.173**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``)
+- **00:00.168**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tvmc.py` (``micro_tvmc.py``)
+- **00:00.168**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)
diff --git a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
index e6b54bf..abbfedb 100644
--- a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
@@ -5,8 +5,8 @@
Computation times
=================
-**00:04.646** total execution time for **how_to_work_with_relay** files:
+**00:04.552** total execution time for **how_to_work_with_relay** files:
-- **00:02.572**: :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)
-- **00:01.855**: :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)
-- **00:00.219**: :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``)
+- **00:02.495**: :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)
+- **00:01.874**: :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)
+- **00:00.183**: :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``)
diff --git a/docs/_sources/how_to/work_with_schedules/schedule_primitives.rst.txt b/docs/_sources/how_to/work_with_schedules/schedule_primitives.rst.txt
index b9b7ead..ae51c73 100644
--- a/docs/_sources/how_to/work_with_schedules/schedule_primitives.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/schedule_primitives.rst.txt
@@ -458,13 +458,13 @@ of computation of `C`.
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {C: Buffer(C_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
- A: Buffer(A_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
+ buffers = {A: Buffer(A_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
+ C: Buffer(C_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
B: Buffer(B_2: Pointer(float32), float32, [(stride_2: int32*m)], [], type="auto")}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (i: int32, 0, m) {
- B[(i*stride_2)] = (A[(i*stride_1)] + 1f32)
- C[(i*stride)] = (B[(i*stride_2)]*2f32)
+ B[(i*stride_2)] = (A[(i*stride)] + 1f32)
+ C[(i*stride_1)] = (B[(i*stride_2)]*2f32)
}
}
@@ -501,12 +501,12 @@ tensor is required.
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A: Buffer(A_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
- C: Buffer(C_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
+ A: Buffer(A_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
B: Buffer(B_2: Pointer(float32), float32, [(stride_2: int32*m)], [], type="auto")}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (i: int32, 0, m) {
- C[(i*stride_1)] = ((A[(i*stride)] + 1f32)*2f32)
+ C[(i*stride)] = ((A[(i*stride_1)] + 1f32)*2f32)
}
}
diff --git a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
index 6dcb06b..67126d5 100644
--- a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
@@ -5,13 +5,13 @@
Computation times
=================
-**00:06.155** total execution time for **how_to_work_with_schedules** files:
+**00:05.815** total execution time for **how_to_work_with_schedules** files:
-- **00:02.225**: :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)
-- **00:01.355**: :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)
-- **00:00.770**: :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)
-- **00:00.766**: :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)
-- **00:00.330**: :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)
-- **00:00.244**: :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``)
-- **00:00.240**: :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)
-- **00:00.225**: :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)
+- **00:02.190**: :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)
+- **00:01.277**: :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)
+- **00:00.731**: :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)
+- **00:00.725**: :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)
+- **00:00.287**: :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)
+- **00:00.208**: :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)
+- **00:00.207**: :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``)
+- **00:00.190**: :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)
diff --git a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
index 28c5969..bc65c12 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -311,12 +311,12 @@ The importing needs to happen before the tensorized GEMV being executed.
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A: Buffer(A_2: Pointer(float32), float32, [65536], []),
- C: Buffer(C_2: Pointer(float32), float32, [524288], []),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [524288], []),
+ A: Buffer(A_2: Pointer(float32), float32, [65536], []),
B: Buffer(B_2: Pointer(float32), float32, [32768], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpmr8tne_a/input0.cc'
- source_filename = "/tmp/tmpmr8tne_a/input0.cc"
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpdnbja0_x/input0.cc'
+ source_filename = "/tmp/tmpdnbja0_x/input0.cc"
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-pc-linux-gnu"
diff --git a/docs/_sources/how_to/work_with_schedules/tuple_inputs.rst.txt b/docs/_sources/how_to/work_with_schedules/tuple_inputs.rst.txt
index 4399c31..329f7aa 100644
--- a/docs/_sources/how_to/work_with_schedules/tuple_inputs.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tuple_inputs.rst.txt
@@ -135,17 +135,17 @@ with :py:func:`te.comm_reducer` as below:
@main = primfn(idx_1: handle, val_1: handle, T_2: handle, T_3: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {T_1: Buffer(T_4: Pointer(int32), int32, [(stride: int32*m: int32)], [], type="auto"),
- T: Buffer(T_5: Pointer(int32), int32, [(stride_1: int32*m)], [], type="auto"),
- idx: Buffer(idx_2: Pointer(int32), int32, [(stride_2: int32*m)], [], type="auto"),
+ buffers = {T: Buffer(T_4: Pointer(int32), int32, [(stride: int32*m: int32)], [], type="auto"),
+ idx: Buffer(idx_2: Pointer(int32), int32, [(stride_1: int32*m)], [], type="auto"),
+ T_1: Buffer(T_5: Pointer(int32), int32, [(stride_2: int32*m)], [], type="auto"),
val: Buffer(val_2: Pointer(int32), int32, [(stride_3: int32*m)], [], type="auto")}
buffer_map = {idx_1: idx, val_1: val, T_2: T, T_3: T_1} {
for (i: int32, 0, m) {
- T[(i*stride_1)] = -1
- T_1[(i*stride)] = -2147483648
+ T[(i*stride)] = -1
+ T_1[(i*stride_2)] = -2147483648
for (k: int32, 0, n: int32) {
- T[(i*stride_1)] = @tir.if_then_else((val[((i*stride_3) + (k*stride_4: int32))] <= T_1[(i*stride)]), T[(i*stride_1)], idx[((i*stride_2) + (k*stride_5: int32))], dtype=int32)
- T_1[(i*stride)] = @tir.if_then_else((val[((i*stride_3) + (k*stride_4))] <= T_1[(i*stride)]), T_1[(i*stride)], val[((i*stride_3) + (k*stride_4))], dtype=int32)
+ T[(i*stride)] = @tir.if_then_else((val[((i*stride_3) + (k*stride_4: int32))] <= T_1[(i*stride_2)]), T[(i*stride)], idx[((i*stride_1) + (k*stride_5: int32))], dtype=int32)
+ T_1[(i*stride_2)] = @tir.if_then_else((val[((i*stride_3) + (k*stride_4))] <= T_1[(i*stride_2)]), T_1[(i*stride_2)], val[((i*stride_3) + (k*stride_4))], dtype=int32)
}
}
}
@@ -193,19 +193,19 @@ in terms of operation.
@main = primfn(A0_1: handle, A1_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A0: Buffer(A0_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
- C: Buffer(C_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
+ A0: Buffer(A0_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
A1: Buffer(A1_2: Pointer(float32), float32, [(stride_2: int32*m)], [], type="auto")}
buffer_map = {A0_1: A0, A1_1: A1, C_1: C} {
allocate(B.v0: Pointer(global float32), float32, [n: int32]), storage_scope = global;
allocate(B.v1: Pointer(global float32), float32, [n]), storage_scope = global;
for (i: int32, 0, m) {
for (j: int32, 0, n) {
- B.v0_1: Buffer(B.v0, float32, [n], [])[j] = (A0[((i*stride) + (j*stride_3: int32))] + 2f32)
- B.v1_1: Buffer(B.v1, float32, [n], [])[j] = (A0[((i*stride) + (j*stride_3))]*3f32)
+ B.v0_1: Buffer(B.v0, float32, [n], [])[j] = (A0[((i*stride_1) + (j*stride_3: int32))] + 2f32)
+ B.v1_1: Buffer(B.v1, float32, [n], [])[j] = (A0[((i*stride_1) + (j*stride_3))]*3f32)
}
for (j_1: int32, 0, n) {
- C[((i*stride_1) + (j_1*stride_4: int32))] = (A1[((i*stride_2) + (j_1*stride_5: int32))] + B.v0_1[j_1])
+ C[((i*stride) + (j_1*stride_4: int32))] = (A1[((i*stride_2) + (j_1*stride_5: int32))] + B.v0_1[j_1])
}
}
}
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index 2f4a50a..6a75595 100644
--- a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
@@ -5,7 +5,7 @@
Computation times
=================
-**00:20.768** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:19.421** total execution time for **topic_vta_tutorials_autotvm** files:
-- **00:20.557**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``)
-- **00:00.211**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)
+- **00:19.245**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``)
+- **00:00.177**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
index a0d8f30..64e4659 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -265,7 +265,7 @@ The compilation steps are:
DeprecationWarning,
/workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the new recommended usage.
relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
- resnet18_v1 inference graph built in 20.80s!
+ resnet18_v1 inference graph built in 20.18s!
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
index 96568e8..c58063c 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -301,7 +301,7 @@ The compilation steps are:
/workspace/python/tvm/relay/build_module.py:439: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 14.62s!
+ yolov3-tiny inference graph built in 14.20s!
diff --git a/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
index f889d0a..f9387d3 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
@@ -5,7 +5,7 @@
Computation times
=================
-**01:27.591** total execution time for **topic_vta_tutorials_frontend** files:
+**01:26.900** total execution time for **topic_vta_tutorials_frontend** files:
-- **00:46.614**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)
-- **00:40.977**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``)
+- **00:46.484**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)
+- **00:40.416**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``)
diff --git a/docs/_sources/topic/vta/tutorials/optimize/convolution_opt.rst.txt b/docs/_sources/topic/vta/tutorials/optimize/convolution_opt.rst.txt
index c5b91d9..4286aaf 100644
--- a/docs/_sources/topic/vta/tutorials/optimize/convolution_opt.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/optimize/convolution_opt.rst.txt
@@ -453,8 +453,8 @@ below.
@main = primfn(data_1: handle, kernel_1: handle, res_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {res: Buffer(res_2: Pointer(int8), int8, [50176], []),
- data: Buffer(data_2: Pointer(int8), int8, [50176], []),
+ buffers = {data: Buffer(data_2: Pointer(int8), int8, [50176], []),
+ res: Buffer(res_2: Pointer(int8), int8, [50176], []),
kernel: Buffer(kernel_2: Pointer(int8), int8, [589824], [])}
buffer_map = {data_1: data, kernel_1: kernel, res_1: res} {
allocate(data_buf: Pointer(global int8), int8, [65536]), storage_scope = global;
diff --git a/docs/_sources/topic/vta/tutorials/optimize/matrix_multiply_opt.rst.txt b/docs/_sources/topic/vta/tutorials/optimize/matrix_multiply_opt.rst.txt
index b15e627..d7be01e 100644
--- a/docs/_sources/topic/vta/tutorials/optimize/matrix_multiply_opt.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/optimize/matrix_multiply_opt.rst.txt
@@ -355,8 +355,8 @@ below:
@main = primfn(data_1: handle, weight_1: handle, res_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {data: Buffer(data_2: Pointer(int8), int8, [1024], []),
- res: Buffer(res_2: Pointer(int8), int8, [1024], []),
+ buffers = {res: Buffer(res_2: Pointer(int8), int8, [1024], []),
+ data: Buffer(data_2: Pointer(int8), int8, [1024], []),
weight: Buffer(weight_2: Pointer(int8), int8, [1048576], [])}
buffer_map = {data_1: data, weight_1: weight, res_1: res} {
allocate(data_buf: Pointer(global int8), int8, [1024]), storage_scope = global;
diff --git a/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
index 94fe3fd..ed4ba68 100644
--- a/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
@@ -5,7 +5,7 @@
Computation times
=================
-**00:03.694** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.563** total execution time for **topic_vta_tutorials_optimize** files:
-- **00:03.107**: :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)
-- **00:00.586**: :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``)
+- **00:03.012**: :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)
+- **00:00.550**: :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``)
diff --git a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
index 4c5b0ab..c49744c 100644
--- a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
@@ -5,7 +5,7 @@
Computation times
=================
-**00:01.001** total execution time for **topic_vta_tutorials** files:
+**00:01.013** total execution time for **topic_vta_tutorials** files:
-- **00:00.506**: :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``)
-- **00:00.495**: :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``)
+- **00:00.508**: :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``)
+- **00:00.506**: :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``)
diff --git a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
index b742a5a..783d448 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -305,7 +305,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 92.564 ms
+ Execution time of this operator: 94.993 ms
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index df3edf8..64ae085 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -268,7 +268,7 @@ standard deviation.
.. code-block:: none
- {'mean': 491.39024845000677, 'median': 491.3854500499838, 'std': 1.1300138324741626}
+ {'mean': 489.56825265006046, 'median': 489.47840924993216, 'std': 0.5823560722188575}
@@ -482,29 +482,30 @@ the tuning data to.
.. code-block:: none
-
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 1/25] Current/Best: 16.59/ 19.72 GFLOPS | Progress: (4/10) | 6.62 s
[Task 1/25] Current/Best: 8.71/ 24.21 GFLOPS | Progress: (8/10) | 9.03 s
[Task 1/25] Current/Best: 14.61/ 24.21 GFLOPS | Progress: (10/10) | 10.26 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 2/25] Current/Best: 6.48/ 23.19 GFLOPS | Progress: (4/10) | 2.38 s
[Task 2/25] Current/Best: 15.35/ 23.19 GFLOPS | Progress: (8/10) | 4.00 s
[Task 2/25] Current/Best: 16.23/ 23.19 GFLOPS | Progress: (10/10) | 4.52 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 3/25] Current/Best: 10.40/ 17.90 GFLOPS | Progress: (4/10) | 3.11 s
[Task 3/25] Current/Best: 10.60/ 19.79 GFLOPS | Progress: (8/10) | 4.71 s
[Task 3/25] Current/Best: 6.46/ 19.79 GFLOPS | Progress: (10/10) | 5.75 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 4/25] Current/Best: 10.00/ 14.78 GFLOPS | Progress: (4/10) | 5.94 s
[Task 4/25] Current/Best: 4.83/ 15.69 GFLOPS | Progress: (8/10) | 8.00 s
[Task 4/25] Current/Best: 6.73/ 15.69 GFLOPS | Progress: (10/10) | 9.03 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 5/25] Current/Best: 1.80/ 13.93 GFLOPS | Progress: (4/10) | 3.34 s
[Task 5/25] Current/Best: 15.89/ 17.56 GFLOPS | Progress: (8/10) | 5.12 s
[Task 5/25] Current/Best: 18.71/ 18.71 GFLOPS | Progress: (10/10) | 5.74 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 6/25] Current/Best: 12.13/ 12.13 GFLOPS | Progress: (4/10) | 4.49 s
[Task 6/25] Current/Best: 12.37/ 20.90 GFLOPS | Progress: (8/10) | 8.33 s
[Task 6/25] Current/Best: 10.21/ 20.90 GFLOPS | Progress: (10/10) | 9.56 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 7/25] Current/Best: 14.03/ 14.03 GFLOPS | Progress: (4/10) | 5.09 s
[Task 7/25] Current/Best: 13.51/ 19.35 GFLOPS | Progress: (8/10) | 6.94 s
[Task 7/25] Current/Best: 11.50/ 19.35 GFLOPS | Progress: (10/10) | 7.81 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 8/25] Current/Best: 5.31/ 14.41 GFLOPS | Progress: (4/10) | 3.91 s
[Task 8/25] Current/Best: 13.50/ 14.41 GFLOPS | Progress: (8/10) | 6.92 s
[Task 8/25] Current/Best: 10.07/ 14.41 GFLOPS | Progress: (10/10) | 7.95 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 9/25] Current/Best: 18.89/ 18.89 GFLOPS | Progress: (4/10) | 12.79 s
[Task 9/25] Current/Best: 12.54/ 18.89 GFLOPS | Progress: (8/10) | 15.36 s
[Task 9/25] Current/Best: 19.53/ 19.53 GFLOPS | Progress: (10/10) | 15.98 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 10/25] Current/Best: 16.60/ 16.60 GFLOPS | Progress: (4/10) | 2.44 s
[Task 10/25] Current/Best: 5.55/ 16.60 GFLOPS | Progress: (8/10) | 4.54 s
[Task 10/25] Current/Best: 3.02/ 16.60 GFLOPS | Progress: (10/10) | 5.66 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 11/25] Current/Best: 16.64/ 24.25 GFLOPS | Progress: (4/10) | 3.43 s
[Task 11/25] Current/Best: 15.92/ 24.25 GFLOPS | Progress: (8/10) | 5.04 s
[Task 11/25] Current/Best: 7.66/ 24.25 GFLOPS | Progress: (10/10) | 6.47 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 12/25] Current/Best: 16.28/ 16.28 GFLOPS | Progress: (4/10) | 2.89 s
[Task 12/25] Current/Best: 6.13/ 16.28 GFLOPS | Progress: (8/10) | 5.21 s
[Task 12/25] Current/Best: 15.14/ 16.28 GFLOPS | Progress: (10/10) | 6.30 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 13/25] Current/Best: 5.59/ 18.23 GFLOPS | Progress: (4/10) | 4.28 s
[Task 13/25] Current/Best: 7.01/ 18.23 GFLOPS | Progress: (8/10) | 8.00 s
[Task 13/25] Current/Best: 18.35/ 18.35 GFLOPS | Progress: (10/10) | 9.94 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 14/25] Current/Best: 12.10/ 19.33 GFLOPS | Progress: (4/10) | 2.75 s Done.
-
[Task 14/25] Current/Best: 7.03/ 22.96 GFLOPS | Progress: (8/10) | 4.67 s
[Task 14/25] Current/Best: 6.45/ 22.96 GFLOPS | Progress: (10/10) | 5.79 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 15/25] Current/Best: 19.92/ 19.92 GFLOPS | Progress: (4/10) | 8.23 s
[Task 15/25] Current/Best: 16.07/ 19.92 GFLOPS | Progress: (8/10) | 10.46 s
[Task 15/25] Current/Best: 21.98/ 21.98 GFLOPS | Progress: (10/10) | 11.00 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 16/25] Current/Best: 19.37/ 19.37 GFLOPS | Progress: (4/10) | 2.99 s
[Task 16/25] Current/Best: 3.09/ 19.37 GFLOPS | Progress: (8/10) | 5.24 s
[Task 16/25] Current/Best: 15.73/ 19.37 GFLOPS | Progress: (10/10) | 6.00 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 17/25] Current/Best: 14.95/ 19.05 GFLOPS | Progress: (4/10) | 3.28 s
[Task 17/25] Current/Best: 10.16/ 19.05 GFLOPS | Progress: (8/10) | 5.93 s
[Task 17/25] Current/Best: 15.91/ 19.05 GFLOPS | Progress: (10/10) | 6.80 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 18/25] Current/Best: 18.66/ 21.87 GFLOPS | Progress: (4/10) | 2.61 s
[Task 18/25] Current/Best: 18.28/ 22.39 GFLOPS | Progress: (8/10) | 4.23 s
[Task 18/25] Current/Best: 9.77/ 22.39 GFLOPS | Progress: (10/10) | 8.23 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 19/25] Current/Best: 24.07/ 24.07 GFLOPS | Progress: (4/10) | 4.21 s
[Task 19/25] Current/Best: 12.72/ 24.07 GFLOPS | Progress: (8/10) | 7.59 s
[Task 19/25] Current/Best: 14.15/ 24.07 GFLOPS | Progress: (10/10) | 8.47 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 20/25] Current/Best: 5.33/ 19.46 GFLOPS | Progress: (4/10) | 3.67 s
[Task 20/25] Current/Best: 18.59/ 19.46 GFLOPS | Progress: (8/10) | 6.51 s
[Task 20/25] Current/Best: 10.74/ 19.46 GFLOPS | Progress: (10/10) | 7.48 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 21/25] Current/Best: 17.19/ 17.19 GFLOPS | Progress: (4/10) | 2.97 s
[Task 21/25] Current/Best: 9.14/ 17.19 GFLOPS | Progress: (8/10) | 5.76 s Done.
- Done.
-
[Task 21/25] Current/Best: 2.62/ 17.19 GFLOPS | Progress: (10/10) | 7.20 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 22/25] Current/Best: 18.09/ 18.09 GFLOPS | Progress: (4/10) | 3.31 s
[Task 22/25] Current/Best: 10.16/ 18.09 GFLOPS | Progress: (8/10) | 6.01 s
[Task 22/25] Current/Best: 7.34/ 18.09 GFLOPS | Progress: (10/10) | 6.92 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 23/25] Current/Best: 12.15/ 18.24 GFLOPS | Progress: (4/10) | 3.84 s
[Task 23/25] Current/Best: 13.76/ 18.39 GFLOPS | Progress: (8/10) | 6.53 s
[Task 23/25] Current/Best: 11.21/ 18.39 GFLOPS | Progress: (10/10) | 8.34 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 24/25] Current/Best: 4.82/ 4.82 GFLOPS | Progress: (4/10) | 1514.50 s
[Task 24/25] Current/Best: 3.36/ 7.91 GFLOPS | Progress: (8/10) | 1540.02 s
[Task 24/25] Current/Best: 1.53/ 7.91 GFLOPS | Progress: (10/10) | 1550.12 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
-
[Task 25/25] Current/Best: 8.56/ 8.96 GFLOPS | Progress: (4/10) | 29.42 s
[Task 25/25] Current/Best: 9.53/ 9.53 GFLOPS | Progress: (8/10) | 42.32 s
[Task 25/25] Current/Best: 7.92/ 9.53 GFLOPS | Progress: (10/10) | 42.82 s
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 1/25] Current/Best: 23.77/ 23.77 GFLOPS | Progress: (4/10) | 6.24 s
[Task 1/25] Current/Best: 16.25/ 23.77 GFLOPS | Progress: (8/10) | 9.03 s
[Task 1/25] Current/Best: 8.88/ 23.77 GFLOPS | Progress: (10/10) | 11.47 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 2/25] Current/Best: 12.03/ 16.92 GFLOPS | Progress: (4/10) | 2.65 s
[Task 2/25] Current/Best: 13.25/ 16.92 GFLOPS | Progress: (8/10) | 4.05 s
[Task 2/25] Current/Best: 20.91/ 20.91 GFLOPS | Progress: (10/10) | 4.53 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 3/25] Current/Best: 6.24/ 20.83 GFLOPS | Progress: (4/10) | 3.56 s
[Task 3/25] Current/Best: 18.58/ 20.83 GFLOPS | Progress: (8/10) | 5.58 s
[Task 3/25] Current/Best: 14.08/ 20.83 GFLOPS | Progress: (10/10) | 6.56 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 4/25] Current/Best: 13.83/ 16.57 GFLOPS | Progress: (4/10) | 2.76 s
[Task 4/25] Current/Best: 6.58/ 18.38 GFLOPS | Progress: (8/10) | 4.26 s
[Task 4/25] Current/Best: 13.42/ 18.38 GFLOPS | Progress: (10/10) | 5.26 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 5/25] Current/Best: 13.08/ 23.57 GFLOPS | Progress: (4/10) | 4.11 s
[Task 5/25] Current/Best: 6.50/ 23.57 GFLOPS | Progress: (8/10) | 6.43 s
[Task 5/25] Current/Best: 9.91/ 23.57 GFLOPS | Progress: (10/10) | 7.61 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 6/25] Current/Best: 5.81/ 16.23 GFLOPS | Progress: (4/10) | 3.73 s
[Task 6/25] Current/Best: 10.74/ 22.71 GFLOPS | Progress: (8/10) | 5.70 s
[Task 6/25] Current/Best: 3.56/ 22.71 GFLOPS | Progress: (10/10) | 7.21 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 7/25] Current/Best: 12.42/ 13.75 GFLOPS | Progress: (4/10) | 4.09 s
[Task 7/25] Current/Best: 19.39/ 19.39 GFLOPS | Progress: (8/10) | 5.80 s
[Task 7/25] Current/Best: 8.48/ 22.51 GFLOPS | Progress: (10/10) | 7.24 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 8/25] Current/Best: 16.33/ 18.75 GFLOPS | Progress: (4/10) | 3.83 s
[Task 8/25] Current/Best: 5.34/ 18.75 GFLOPS | Progress: (8/10) | 7.19 s
[Task 8/25] Current/Best: 14.13/ 18.75 GFLOPS | Progress: (10/10) | 9.12 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 9/25] Current/Best: 13.02/ 21.21 GFLOPS | Progress: (4/10) | 3.70 s
[Task 9/25] Current/Best: 5.13/ 21.21 GFLOPS | Progress: (8/10) | 9.15 s
[Task 9/25] Current/Best: 15.37/ 21.21 GFLOPS | Progress: (10/10) | 9.90 s Done.
+
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 10/25] Current/Best: 1.56/ 20.84 GFLOPS | Progress: (4/10) | 3.22 s
[Task 10/25] Current/Best: 7.77/ 20.84 GFLOPS | Progress: (8/10) | 5.50 s
[Task 10/25] Current/Best: 6.91/ 20.84 GFLOPS | Progress: (10/10) | 6.25 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 11/25] Current/Best: 21.16/ 21.16 GFLOPS | Progress: (4/10) | 3.86 s
[Task 11/25] Current/Best: 17.57/ 22.07 GFLOPS | Progress: (8/10) | 5.75 s
[Task 11/25] Current/Best: 12.78/ 22.07 GFLOPS | Progress: (10/10) | 7.45 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 12/25] Current/Best: 18.03/ 18.03 GFLOPS | Progress: (4/10) | 7.77 s
[Task 12/25] Current/Best: 18.14/ 18.35 GFLOPS | Progress: (8/10) | 9.90 s
[Task 12/25] Current/Best: 12.55/ 18.35 GFLOPS | Progress: (10/10) | 10.95 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 13/25] Current/Best: 12.23/ 15.69 GFLOPS | Progress: (4/10) | 3.65 s
[Task 13/25] Current/Best: 11.38/ 20.38 GFLOPS | Progress: (8/10) | 5.92 s
[Task 13/25] Current/Best: 1.57/ 20.38 GFLOPS | Progress: (10/10) | 8.70 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 14/25] Current/Best: 16.06/ 16.06 GFLOPS | Progress: (4/10) | 4.82 s
[Task 14/25] Current/Best: 6.33/ 16.06 GFLOPS | Progress: (8/10) | 7.03 s
[Task 14/25] Current/Best: 12.93/ 17.27 GFLOPS | Progress: (10/10) | 8.13 s Done.
+
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 15/25] Current/Best: 13.37/ 19.47 GFLOPS | Progress: (4/10) | 2.91 s
[Task 15/25] Current/Best: 19.31/ 19.47 GFLOPS | Progress: (8/10) | 5.95 s
[Task 15/25] Current/Best: 15.19/ 19.47 GFLOPS | Progress: (10/10) | 6.58 s Done.
+
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 16/25] Current/Best: 11.39/ 17.03 GFLOPS | Progress: (4/10) | 3.11 s
[Task 16/25] Current/Best: 13.97/ 17.03 GFLOPS | Progress: (8/10) | 5.00 s
[Task 16/25] Current/Best: 11.83/ 17.03 GFLOPS | Progress: (10/10) | 6.36 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 17/25] Current/Best: 6.23/ 20.67 GFLOPS | Progress: (4/10) | 3.54 s
[Task 17/25] Current/Best: 11.09/ 20.67 GFLOPS | Progress: (8/10) | 6.84 s
[Task 17/25] Current/Best: 8.77/ 20.67 GFLOPS | Progress: (10/10) | 7.90 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 18/25] Current/Best: 14.09/ 18.65 GFLOPS | Progress: (4/10) | 2.87 s
[Task 18/25] Current/Best: 11.91/ 22.01 GFLOPS | Progress: (8/10) | 4.80 s
[Task 18/25] Current/Best: 3.04/ 22.01 GFLOPS | Progress: (10/10) | 6.58 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 19/25] Current/Best: 6.99/ 22.58 GFLOPS | Progress: (4/10) | 4.25 s
[Task 19/25] Current/Best: 19.17/ 22.58 GFLOPS | Progress: (8/10) | 7.21 s
[Task 19/25] Current/Best: 6.19/ 22.58 GFLOPS | Progress: (10/10) | 9.86 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 20/25] Current/Best: 16.01/ 18.06 GFLOPS | Progress: (4/10) | 2.83 s
[Task 20/25] Current/Best: 21.65/ 21.65 GFLOPS | Progress: (8/10) | 5.21 s
[Task 20/25] Current/Best: 9.27/ 21.65 GFLOPS | Progress: (10/10) | 7.17 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
+
[Task 21/25] Current/Best: 15.89/ 16.94 GFLOPS | Progress: (4/10) | 2.50 s
[Task 21/25] Current/Best: 20.35/ 20.35 GFLOPS | Progress: (8/10) | 4.28 s
[Task 21/25] Current/Best: 10.12/ 20.35 GFLOPS | Progress: (10/10) | 4.86 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 22/25] Current/Best: 10.58/ 17.52 GFLOPS | Progress: (4/10) | 4.50 s
[Task 22/25] Current/Best: 13.48/ 21.42 GFLOPS | Progress: (8/10) | 5.88 s
[Task 22/25] Current/Best: 2.54/ 21.42 GFLOPS | Progress: (10/10) | 6.91 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 23/25] Current/Best: 14.15/ 23.24 GFLOPS | Progress: (4/10) | 3.73 s
[Task 23/25] Current/Best: 11.22/ 23.24 GFLOPS | Progress: (8/10) | 5.90 s
[Task 23/25] Current/Best: 6.20/ 23.24 GFLOPS | Progress: (10/10) | 12.20 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 24/25] Current/Best: 3.00/ 3.80 GFLOPS | Progress: (4/10) | 7.81 s
[Task 24/25] Current/Best: 5.96/ 8.13 GFLOPS | Progress: (8/10) | 9.13 s
[Task 24/25] Current/Best: 5.61/ 8.13 GFLOPS | Progress: (10/10) | 9.82 s Done.
+
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 25/25] Current/Best: 1.55/ 5.27 GFLOPS | Progress: (4/10) | 32.16 s
[Task 25/25] Current/Best: 4.54/ 8.54 GFLOPS | Progress: (8/10) | 49.70 s
[Task 25/25] Current/Best: 5.92/ 8.54 GFLOPS | Progress: (10/10) | 61.99 s
The output from this tuning process will look something like this:
@@ -592,7 +593,7 @@ Verify that the optimized model runs and produces the same results:
.. code-block:: none
- class='n02123045 tabby, tabby cat' with probability=0.610552
+ class='n02123045 tabby, tabby cat' with probability=0.610551
class='n02123159 tiger cat' with probability=0.367180
class='n02124075 Egyptian cat' with probability=0.019365
class='n02129604 tiger, Panthera tigris' with probability=0.001273
@@ -646,8 +647,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 450.5758023400267, 'median': 450.24263310006063, 'std': 1.294131255260828}
- unoptimized: {'mean': 491.39024845000677, 'median': 491.3854500499838, 'std': 1.1300138324741626}
+ optimized: {'mean': 439.65931273000024, 'median': 438.314017850098, 'std': 2.746597250938707}
+ unoptimized: {'mean': 489.56825265006046, 'median': 489.47840924993216, 'std': 0.5823560722188575}
@@ -667,7 +668,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 32 minutes 41.859 seconds)
+ **Total running time of the script:** ( 7 minutes 17.563 seconds)
.. _sphx_glr_download_tutorial_autotvm_relay_x86.py:
diff --git a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
index d13821f..4236614 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -235,7 +235,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.243e-07 secs/op
+ 1.256e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 2fb1239..5d8fc66 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -230,7 +230,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0x2092de30)), stage(b, placeholder(b, 0x22703940)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(mi [...]
+ [stage(a, placeholder(a, 0xcfb8f30)), stage(b, placeholder(b, 0x918fcf0)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min= [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index e9131a3..61ea039 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,17 +5,17 @@
Computation times
=================
-**35:14.951** total execution time for **tutorial** files:
+**09:56.984** total execution time for **tutorial** files:
-- **32:41.859**: :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)
-- **00:58.426**: :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)
-- **00:51.099**: :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``)
-- **00:25.416**: :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)
-- **00:16.582**: :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)
-- **00:00.700**: :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)
-- **00:00.551**: :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)
-- **00:00.176**: :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``)
-- **00:00.037**: :ref:`sphx_glr_tutorial_install.py` (``install.py``)
-- **00:00.036**: :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``)
-- **00:00.035**: :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)
-- **00:00.035**: :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``)
+- **07:17.563**: :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)
+- **01:00.867**: :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)
+- **00:53.898**: :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``)
+- **00:25.182**: :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)
+- **00:17.949**: :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)
+- **00:00.692**: :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)
+- **00:00.549**: :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)
+- **00:00.171**: :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``)
+- **00:00.030**: :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``)
+- **00:00.028**: :ref:`sphx_glr_tutorial_install.py` (``install.py``)
+- **00:00.027**: :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``)
+- **00:00.027**: :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index 7a7b66f..a2d8492 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -243,7 +243,7 @@ helper function to run a profile of the TVM generated code.
.. code-block:: none
- Numpy running time: 0.000007
+ Numpy running time: 0.000008
naive: 0.000006
@@ -387,18 +387,18 @@ factor to be the number of threads on your CPU.
.. code-block:: none
- vector: 0.000029
+ vector: 0.000024
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A: Buffer(A_2: Pointer(float32), float32, [(stride: int32*n: int32)], [], type="auto"),
- C: Buffer(C_2: Pointer(float32), float32, [(stride_1: int32*n)], [], type="auto"),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [(stride: int32*n: int32)], [], type="auto"),
+ A: Buffer(A_2: Pointer(float32), float32, [(stride_1: int32*n)], [], type="auto"),
B: Buffer(B_2: Pointer(float32), float32, [(stride_2: int32*n)], [], type="auto")}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (i.outer: int32, 0, floordiv((n + 3), 4)) "parallel" {
for (i.inner.s: int32, 0, 4) {
if @tir.likely((((i.outer*4) + i.inner.s) < n), dtype=bool) {
let cse_var_1: int32 = ((i.outer*4) + i.inner.s)
- C[(cse_var_1*stride_1)] = (A[(cse_var_1*stride)] + B[(cse_var_1*stride_2)])
+ C[(cse_var_1*stride)] = (A[(cse_var_1*stride_1)] + B[(cse_var_1*stride_2)])
}
}
}
@@ -436,10 +436,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 6.735409988323226e-06 1.0
- naive 5.823300000000001e-06 0.8645798860196342
- parallel 6.0327e-06 0.8956693075044471
- vector 2.8760200000000003e-05 4.269999903474298
+ numpy 7.593640002596658e-06 1.0
+ naive 5.819499999999999e-06 0.7663650104574369
+ parallel 6.0047e-06 0.7907538411021174
+ vector 2.4472e-05 3.2226968873467476
@@ -828,7 +828,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.018702
+ Numpy running time: 0.017699
@@ -884,7 +884,7 @@ optimizations.
.. code-block:: none
- none: 3.244222
+ none: 3.434988
@@ -911,8 +911,8 @@ nested loops over the indices of the A and B matrices.
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
- C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
+ A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
B: Buffer(B_2: Pointer(float32), float32, [1048576], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (x: int32, 0, 1024) {
@@ -982,7 +982,7 @@ schedule.
.. code-block:: none
- blocking: 0.285139
+ blocking: 0.295816
@@ -1073,7 +1073,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.328088
+ vectorization: 0.330708
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
@@ -1144,7 +1144,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.116875
+ loop permutation: 0.114240
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
@@ -1240,11 +1240,11 @@ optimized schedule.
.. code-block:: none
- array packing: 0.110480
+ array packing: 0.108623
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
- A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
B: Buffer(B_2: Pointer(float32), float32, [1048576], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
allocate(packedB: Pointer(global float32x32), float32x32, [32768]), storage_scope = global {
@@ -1330,7 +1330,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.110613
+ block caching: 0.110040
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
@@ -1413,7 +1413,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.144371
+ parallelization: 0.144407
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
@@ -1491,13 +1491,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.2442223216999997 1.0
- blocking 0.2851393656 0.08789143817079238
- vectorization 0.328087547 0.10112979767307666
- loop permutation 0.1168754399 0.03602571843435079
- array packing 0.1104800113 0.03405438972570399
- block caching 0.11061262320000001 0.03409526605502118
- parallelization 0.1443707233 0.04450087231517122
+ none 3.4349884542 1.0
+ blocking 0.2958159905 0.08611848174869478
+ vectorization 0.3307081293 0.0962763437809054
+ loop permutation 0.1142404741 0.03325789172895672
+ array packing 0.10862342900000002 0.03162264748435614
+ block caching 0.110040035 0.03203505236399056
+ parallelization 0.14440731290000003 0.042040115949569364
@@ -1532,6 +1532,11 @@ operations with tunable parameters that allows you to automatically optimize
the computation for specific platforms.
+.. rst-class:: sphx-glr-timing
+
+ **Total running time of the script:** ( 1 minutes 0.867 seconds)
+
+
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index 43a793d..ae49abe 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-15c1f75b84dde60df3f7f957ab2f1827ff6a76ef
+6eaa5e6f29420b2c85ec6e3c5f83566d60c7d641
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 451ad15..6b55b2d 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -401,7 +401,7 @@
</div>
<img alt="../../_images/sphx_glr_from_mxnet_001.png" class="sphx-glr-single-img" src="../../_images/sphx_glr_from_mxnet_001.png" />
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip1e1ff422-41ea-4070-b3fe-e7ac181ee7fa from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip248dc537-6341-4c0b-a084-22d8eb23d2ad from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
x (1, 3, 224, 224)
</pre></div>
</div>
diff --git a/docs/how_to/compile_models/from_paddle.html b/docs/how_to/compile_models/from_paddle.html
index 67ebf70..43f3b53 100644
--- a/docs/how_to/compile_models/from_paddle.html
+++ b/docs/how_to/compile_models/from_paddle.html
@@ -444,7 +444,7 @@ A quick solution is</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>TVM prediction top-1 id: 282, class name: 282: 'tiger cat',
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 5.840 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 3.428 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-paddle-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/16269b77359771348d507395692524cf/from_paddle.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_paddle.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index 8116f0f..3b2636c 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -387,10 +387,9 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
0%| | 0.00/44.7M [00:00<?, ?B/s]
- 5%|5 | 2.38M/44.7M [00:00<00:01, 24.9MB/s]
- 15%|#4 | 6.66M/44.7M [00:00<00:01, 36.7MB/s]
- 45%|####4 | 20.1M/44.7M [00:00<00:00, 84.2MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 117MB/s]
+ 29%|##9 | 13.1M/44.7M [00:00<00:00, 137MB/s]
+ 69%|######9 | 31.0M/44.7M [00:00<00:00, 167MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 173MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_tensorflow.html b/docs/how_to/compile_models/from_tensorflow.html
index 16900ce..5199f60 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -611,7 +611,7 @@ banana (score = 0.00022)
desk (score = 0.00019)
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 4.056 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 7.096 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-tensorflow-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7f1d3d1b878694c201c614c807cdebc8/from_tensorflow.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_tensorflow.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/sg_execution_times.html b/docs/how_to/compile_models/sg_execution_times.html
index 93cb5b3..a219b01 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -300,18 +300,18 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:29.039</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:28.088</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<ul class="simple">
-<li><p><strong>01:05.840</strong>: <a class="reference internal" href="from_paddle.html#sphx-glr-how-to-compile-models-from-paddle-py"><span class="std std-ref">Compile PaddlePaddle Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_paddle.py</span></code>)</p></li>
-<li><p><strong>01:04.056</strong>: <a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></li>
-<li><p><strong>00:56.106</strong>: <a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></li>
-<li><p><strong>00:41.382</strong>: <a class="reference internal" href="from_caffe2.html#sphx-glr-how-to-compile-models-from-caffe2-py"><span class="std std-ref">Compile Caffe2 Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_caffe2.py</span></code>)</p></li>
-<li><p><strong>00:25.177</strong>: <a class="reference internal" href="from_tflite.html#sphx-glr-how-to-compile-models-from-tflite-py"><span class="std std-ref">Compile TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tflite.py</span></code>)</p></li>
-<li><p><strong>00:20.400</strong>: <a class="reference internal" href="from_mxnet.html#sphx-glr-how-to-compile-models-from-mxnet-py"><span class="std std-ref">Compile MXNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_mxnet.py</span></code>)</p></li>
-<li><p><strong>00:20.264</strong>: <a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></li>
-<li><p><strong>00:19.275</strong>: <a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></li>
-<li><p><strong>00:14.105</strong>: <a class="reference internal" href="from_keras.html#sphx-glr-how-to-compile-models-from-keras-py"><span class="std std-ref">Compile Keras Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_keras.py</span></code>)</p></li>
-<li><p><strong>00:02.433</strong>: <a class="reference internal" href="from_onnx.html#sphx-glr-how-to-compile-models-from-onnx-py"><span class="std std-ref">Compile ONNX Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_onnx.py</span></code>)</p></li>
+<li><p><strong>01:07.096</strong>: <a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></li>
+<li><p><strong>01:03.428</strong>: <a class="reference internal" href="from_paddle.html#sphx-glr-how-to-compile-models-from-paddle-py"><span class="std std-ref">Compile PaddlePaddle Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_paddle.py</span></code>)</p></li>
+<li><p><strong>00:54.993</strong>: <a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></li>
+<li><p><strong>00:41.121</strong>: <a class="reference internal" href="from_caffe2.html#sphx-glr-how-to-compile-models-from-caffe2-py"><span class="std std-ref">Compile Caffe2 Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_caffe2.py</span></code>)</p></li>
+<li><p><strong>00:25.700</strong>: <a class="reference internal" href="from_tflite.html#sphx-glr-how-to-compile-models-from-tflite-py"><span class="std std-ref">Compile TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tflite.py</span></code>)</p></li>
+<li><p><strong>00:21.032</strong>: <a class="reference internal" href="from_mxnet.html#sphx-glr-how-to-compile-models-from-mxnet-py"><span class="std std-ref">Compile MXNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_mxnet.py</span></code>)</p></li>
+<li><p><strong>00:20.145</strong>: <a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></li>
+<li><p><strong>00:18.803</strong>: <a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></li>
+<li><p><strong>00:13.426</strong>: <a class="reference internal" href="from_keras.html#sphx-glr-how-to-compile-models-from-keras-py"><span class="std std-ref">Compile Keras Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_keras.py</span></code>)</p></li>
+<li><p><strong>00:02.343</strong>: <a class="reference internal" href="from_onnx.html#sphx-glr-how-to-compile-models-from-onnx-py"><span class="std std-ref">Compile ONNX Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_onnx.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 94e4036..41e9784 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -622,7 +622,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.1042 16.0792 16.7479 15.5151 0.4439
+ 15.5896 15.6136 15.6952 15.4707 0.0816
</pre></div>
</div>
</div>
diff --git a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
index 45d1c13..29bfe85 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -409,13 +409,28 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
0%| | 0.00/170M [00:00<?, ?B/s]
- 11%|#1 | 19.1M/170M [00:00<00:00, 200MB/s]
- 26%|##6 | 44.2M/170M [00:00<00:00, 237MB/s]
- 41%|#### | 69.4M/170M [00:00<00:00, 249MB/s]
- 56%|#####6 | 95.3M/170M [00:00<00:00, 258MB/s]
- 71%|#######1 | 121M/170M [00:00<00:00, 262MB/s]
- 86%|########6 | 146M/170M [00:00<00:00, 264MB/s]
-100%|##########| 170M/170M [00:00<00:00, 257MB/s]
+ 2%|2 | 3.45M/170M [00:00<00:04, 35.9MB/s]
+ 5%|5 | 9.11M/170M [00:00<00:03, 49.3MB/s]
+ 11%|# | 18.2M/170M [00:00<00:02, 69.9MB/s]
+ 15%|#4 | 24.9M/170M [00:00<00:02, 54.5MB/s]
+ 18%|#7 | 30.5M/170M [00:00<00:02, 50.4MB/s]
+ 21%|## | 35.5M/170M [00:00<00:03, 40.6MB/s]
+ 26%|##5 | 43.7M/170M [00:00<00:02, 51.2MB/s]
+ 29%|##9 | 49.8M/170M [00:01<00:02, 54.1MB/s]
+ 33%|###2 | 55.4M/170M [00:01<00:03, 36.3MB/s]
+ 36%|###6 | 61.7M/170M [00:01<00:02, 42.1MB/s]
+ 40%|#### | 68.5M/170M [00:01<00:02, 48.8MB/s]
+ 44%|####3 | 74.1M/170M [00:01<00:02, 49.0MB/s]
+ 50%|####9 | 84.1M/170M [00:01<00:01, 62.3MB/s]
+ 56%|#####6 | 95.2M/170M [00:01<00:01, 76.5MB/s]
+ 64%|######3 | 108M/170M [00:01<00:00, 92.0MB/s]
+ 69%|######9 | 117M/170M [00:02<00:00, 93.5MB/s]
+ 76%|#######5 | 128M/170M [00:02<00:00, 99.9MB/s]
+ 82%|########1 | 139M/170M [00:02<00:00, 102MB/s]
+ 88%|########7 | 149M/170M [00:02<00:00, 85.6MB/s]
+ 93%|#########3| 158M/170M [00:02<00:00, 47.0MB/s]
+ 97%|#########6| 165M/170M [00:02<00:00, 50.6MB/s]
+100%|##########| 170M/170M [00:03<00:00, 59.3MB/s]
/usr/local/lib/python3.6/dist-packages/torch/functional.py:445: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:2157.)
return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]
/usr/local/lib/python3.6/dist-packages/torch/nn/functional.py:3702: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
@@ -510,7 +525,7 @@ torchvision rcnn models.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Get 9 valid boxes
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 59.751 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 57.698 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-object-detection-pytorch-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7795da4b258c8feff986668b95ef57ad/deploy_object_detection_pytorch.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_object_detection_pytorch.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_prequantized.html b/docs/how_to/deploy_models/deploy_prequantized.html
index 41ddc92..66bf2c1 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -450,9 +450,11 @@ training. Other models require a full post training calibration.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
0%| | 0.00/13.6M [00:00<?, ?B/s]
- 42%|####1 | 5.69M/13.6M [00:00<00:00, 59.6MB/s]
- 84%|########3 | 11.4M/13.6M [00:00<00:00, 58.7MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 59.9MB/s]
+ 19%|#9 | 2.59M/13.6M [00:00<00:00, 27.1MB/s]
+ 38%|###8 | 5.17M/13.6M [00:00<00:00, 25.7MB/s]
+ 56%|#####6 | 7.64M/13.6M [00:00<00:00, 25.4MB/s]
+ 82%|########1 | 11.1M/13.6M [00:00<00:00, 29.5MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 30.6MB/s]
</pre></div>
</div>
</div>
@@ -545,7 +547,7 @@ output values are identical out of 1000 outputs from mobilenet v2.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.2507 90.1289 91.9421 89.9283 0.3267
+ 90.4215 90.5000 92.3945 89.8433 0.3587
</pre></div>
</div>
<div class="admonition note">
@@ -584,7 +586,7 @@ This includes support for the VNNI 8 bit dot product instruction (CascadeLake or
<div class="section" id="deploy-a-quantized-tflite-model">
<h2>Deploy a quantized TFLite Model<a class="headerlink" href="#deploy-a-quantized-tflite-model" title="Permalink to this headline">¶</a></h2>
<p>TODO</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 6.844 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 6.068 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/fb8217c13f4351224c6cf3aacf1a87fc/deploy_prequantized.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_prequantized.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_prequantized_tflite.html b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
index cc62b4b..139e5f9 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -540,7 +540,7 @@ TFLite Top-5 labels: [387 102 386 341 880]
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 117.8534 117.8098 122.0296 116.9189 0.6196
+ 116.9473 116.7776 132.1289 115.4894 1.6453
</pre></div>
</div>
<div class="admonition note">
@@ -568,7 +568,7 @@ network for ARM CPU</span></a>.</p></li>
</ul>
</div></blockquote>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 59.810 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 59.056 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-tflite-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/56691c7a27d45da61d112276334640d3/deploy_prequantized_tflite.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_prequantized_tflite.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_quantized.html b/docs/how_to/deploy_models/deploy_quantized.html
index 259ce3c..c8b0a95 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -480,7 +480,7 @@ for calibration. But the accuracy might be impacted.</p>
DeprecationWarning,
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 13.356 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 10.579 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-quantized-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7810ecf51bfc05f7d5e8a400ac3e815d/deploy_quantized.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_quantized.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
index 39c08ad..10d758b 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -415,24 +415,25 @@ to your device.</p>
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
0%| | 0/132723 [00:00<?, ?KB/s]
- 5%|5 | 7086/132723 [00:00<00:01, 70852.47KB/s]
- 12%|#2 | 16073/132723 [00:00<00:01, 82036.16KB/s]
- 19%|#8 | 24678/132723 [00:00<00:01, 83865.46KB/s]
- 25%|##4 | 33065/132723 [00:00<00:01, 60847.69KB/s]
- 30%|### | 40087/132723 [00:00<00:01, 55062.75KB/s]
- 37%|###6 | 48811/132723 [00:00<00:01, 63566.58KB/s]
- 42%|####1 | 55724/132723 [00:00<00:01, 51224.69KB/s]
- 49%|####8 | 64444/132723 [00:01<00:01, 59697.83KB/s]
- 54%|#####3 | 71167/132723 [00:01<00:01, 50614.43KB/s]
- 58%|#####7 | 76917/132723 [00:01<00:01, 49747.80KB/s]
- 62%|######2 | 82353/132723 [00:01<00:01, 45939.26KB/s]
- 69%|######8 | 91049/132723 [00:01<00:00, 55547.15KB/s]
- 74%|#######4 | 98288/132723 [00:01<00:00, 50102.14KB/s]
- 80%|######## | 106818/132723 [00:01<00:00, 58229.84KB/s]
- 85%|########5 | 113458/132723 [00:02<00:00, 47222.48KB/s]
- 90%|########9 | 118872/132723 [00:02<00:00, 38675.19KB/s]
- 96%|#########6| 127652/132723 [00:02<00:00, 48387.95KB/s]
-100%|##########| 132723/132723 [00:02<00:00, 53825.86KB/s]
+ 5%|4 | 6078/132723 [00:00<00:02, 60773.25KB/s]
+ 9%|9 | 12156/132723 [00:00<00:02, 58694.72KB/s]
+ 14%|#3 | 18031/132723 [00:00<00:02, 56270.28KB/s]
+ 19%|#9 | 25568/132723 [00:00<00:01, 63523.50KB/s]
+ 26%|##6 | 34901/132723 [00:00<00:01, 59493.23KB/s]
+ 31%|###1 | 41301/132723 [00:00<00:01, 60628.27KB/s]
+ 37%|###7 | 49140/132723 [00:00<00:01, 58636.55KB/s]
+ 42%|####1 | 55094/132723 [00:00<00:01, 58796.44KB/s]
+ 47%|####7 | 62859/132723 [00:01<00:01, 64039.76KB/s]
+ 52%|#####2 | 69372/132723 [00:01<00:01, 55440.98KB/s]
+ 57%|#####7 | 75840/132723 [00:01<00:00, 57835.80KB/s]
+ 62%|######1 | 81905/132723 [00:01<00:01, 49313.71KB/s]
+ 68%|######8 | 90545/132723 [00:01<00:00, 58407.61KB/s]
+ 74%|#######3 | 98063/132723 [00:01<00:00, 62756.86KB/s]
+ 79%|#######8 | 104719/132723 [00:01<00:00, 51719.67KB/s]
+ 86%|########6 | 114680/132723 [00:01<00:00, 61724.95KB/s]
+ 94%|#########3| 124198/132723 [00:02<00:00, 70079.29KB/s]
+ 99%|#########9| 131771/132723 [00:02<00:00, 64713.43KB/s]
+100%|##########| 132723/132723 [00:02<00:00, 60259.54KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -472,7 +473,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
</pre></div>
</div>
<img alt="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" class="sphx-glr-single-img" src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" />
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 24.540 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 17.502 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-ssd-gluoncv-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/cccb17d28e5e8b2e94ea8cd5ec59f6ed/deploy_ssd_gluoncv.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_ssd_gluoncv.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/sg_execution_times.html b/docs/how_to/deploy_models/sg_execution_times.html
index ec4396b..ac01782 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -300,16 +300,16 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>10:33.678</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>10:19.711</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<ul class="simple">
-<li><p><strong>02:59.751</strong>: <a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></li>
-<li><p><strong>02:24.540</strong>: <a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></li>
-<li><p><strong>01:59.810</strong>: <a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></li>
-<li><p><strong>01:13.356</strong>: <a class="reference internal" href="deploy_quantized.html#sphx-glr-how-to-deploy-models-deploy-quantized-py"><span class="std std-ref">Deploy a Quantized Model on Cuda</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_quantized.py</span></code>)</p></li>
-<li><p><strong>01:06.844</strong>: <a class="reference internal" href="deploy_prequantized.html#sphx-glr-how-to-deploy-models-deploy-prequantized-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized.py</span></code>)</p></li>
-<li><p><strong>00:27.254</strong>: <a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></li>
-<li><p><strong>00:21.924</strong>: <a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></li>
-<li><p><strong>00:00.200</strong>: <a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></li>
+<li><p><strong>02:57.698</strong>: <a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></li>
+<li><p><strong>02:17.502</strong>: <a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></li>
+<li><p><strong>01:59.056</strong>: <a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></li>
+<li><p><strong>01:10.579</strong>: <a class="reference internal" href="deploy_quantized.html#sphx-glr-how-to-deploy-models-deploy-quantized-py"><span class="std std-ref">Deploy a Quantized Model on Cuda</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_quantized.py</span></code>)</p></li>
+<li><p><strong>01:06.068</strong>: <a class="reference internal" href="deploy_prequantized.html#sphx-glr-how-to-deploy-models-deploy-prequantized-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized.py</span></code>)</p></li>
+<li><p><strong>00:26.860</strong>: <a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></li>
+<li><p><strong>00:21.779</strong>: <a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></li>
+<li><p><strong>00:00.169</strong>: <a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/extend_tvm/bring_your_own_datatypes.html b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
index 54c16cf..eb09b73 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -588,7 +588,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip0246fe14-6daf-4336-90cf-efe6bfe45a86 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipecb261bd-611f-4adb-944c-58ada011921f from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
</pre></div>
</div>
<p>It’s easy to execute MobileNet with native TVM:</p>
@@ -650,7 +650,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Check failed: (lower) is false: Intrinsic lowering function for target llvm, intrinsic name tir.sqrt, type 150 not found
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Check failed: (lower) is false: FloatImm lowering function for target llvm type 150 not found
</pre></div>
</div>
<p>When we attempt to run the model, we get a familiar error telling us that more functions need to be registerd for myfloat.</p>
diff --git a/docs/how_to/extend_tvm/sg_execution_times.html b/docs/how_to/extend_tvm/sg_execution_times.html
index e149ace..553aac4 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -300,12 +300,12 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-extend-tvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:38.569</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:37.579</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:34.989</strong>: <a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></li>
-<li><p><strong>00:02.285</strong>: <a class="reference internal" href="use_pass_instrument.html#sphx-glr-how-to-extend-tvm-use-pass-instrument-py"><span class="std std-ref">How to Use TVM Pass Instrument</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_instrument.py</span></code>)</p></li>
-<li><p><strong>00:01.089</strong>: <a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></li>
-<li><p><strong>00:00.206</strong>: <a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></li>
+<li><p><strong>00:33.921</strong>: <a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></li>
+<li><p><strong>00:02.219</strong>: <a class="reference internal" href="use_pass_instrument.html#sphx-glr-how-to-extend-tvm-use-pass-instrument-py"><span class="std std-ref">How to Use TVM Pass Instrument</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_instrument.py</span></code>)</p></li>
+<li><p><strong>00:01.264</strong>: <a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></li>
+<li><p><strong>00:00.175</strong>: <a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index d39c0b8..40ee2aa 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -486,10 +486,10 @@ profile the execution time of each passes.</p>
</div>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 6242us [6242us] (45.59%; 45.59%)
-FoldScaleAxis: 7449us [2us] (54.41%; 54.41%)
- FoldConstant: 7447us [1533us] (54.39%; 99.97%)
- InferType: 5914us [5914us] (43.19%; 79.41%)
+InferType: 6132us [6132us] (45.54%; 45.54%)
+FoldScaleAxis: 7333us [2us] (54.46%; 54.46%)
+ FoldConstant: 7330us [1513us] (54.44%; 99.97%)
+ InferType: 5818us [5818us] (43.21%; 79.37%)
</pre></div>
</div>
</div>
@@ -512,10 +512,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
</div>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 6012us [6012us] (44.82%; 44.82%)
-FoldScaleAxis: 7401us [2us] (55.18%; 55.18%)
- FoldConstant: 7399us [1538us] (55.16%; 99.97%)
- InferType: 5861us [5861us] (43.70%; 79.22%)
+InferType: 5895us [5895us] (44.68%; 44.68%)
+FoldScaleAxis: 7298us [2us] (55.32%; 55.32%)
+ FoldConstant: 7296us [1532us] (55.30%; 99.97%)
+ InferType: 5765us [5765us] (43.69%; 79.00%)
</pre></div>
</div>
<p>Register empty list to clear existing instruments.</p>
diff --git a/docs/how_to/optimize_operators/opt_conv_cuda.html b/docs/how_to/optimize_operators/opt_conv_cuda.html
index 248248a..d7a155e 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -534,7 +534,7 @@ latency of convolution.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 37.764379 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 44.771740 ms
</pre></div>
</div>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-optimize-operators-opt-conv-cuda-py">
diff --git a/docs/how_to/optimize_operators/opt_conv_tensorcore.html b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
index 9a7b90b..9b7bcf3 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -876,7 +876,7 @@ be able to run on our build server</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 7.340276 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 7.085309 ms
</pre></div>
</div>
</div>
diff --git a/docs/how_to/optimize_operators/opt_gemm.html b/docs/how_to/optimize_operators/opt_gemm.html
index 814c193..74bbbba 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -431,8 +431,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019323
-Baseline: 3.247969
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.017687
+Baseline: 3.448237
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -493,7 +493,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.333472
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.291418
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -561,7 +561,7 @@ vastly.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.346588
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.323982
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -571,8 +571,8 @@ vastly.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
- A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
B: Buffer(B_2: Pointer(float32), float32, [1048576], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (m.outer: int32, 0, 32) {
@@ -623,7 +623,7 @@ the access pattern for A matrix is more cache friendly.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.116174
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.116501
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -633,8 +633,8 @@ the access pattern for A matrix is more cache friendly.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
- A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
+ C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
B: Buffer(B_2: Pointer(float32), float32, [1048576], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (m.outer: int32, 0, 32) {
@@ -707,7 +707,7 @@ flattening.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.110871
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.110649
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -794,7 +794,7 @@ write to C when all the block results are ready.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111154
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111518
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -885,7 +885,7 @@ write to C when all the block results are ready.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.144823
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.144946
</pre></div>
</div>
<p>Here is the generated IR after parallelization.</p>
@@ -895,8 +895,8 @@ write to C when all the block results are ready.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
- C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [1048576], []),
+ A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
B: Buffer(B_2: Pointer(float32), float32, [1048576], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
allocate(packedB: Pointer(global float32x32), float32x32, [32768]), storage_scope = global {
diff --git a/docs/how_to/optimize_operators/sg_execution_times.html b/docs/how_to/optimize_operators/sg_execution_times.html
index a91d44f..5a00ff0 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -300,11 +300,11 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:35.235</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:34.744</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:32.549</strong>: <a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></li>
-<li><p><strong>00:01.478</strong>: <a class="reference internal" href="opt_conv_tensorcore.html#sphx-glr-how-to-optimize-operators-opt-conv-tensorcore-py"><span class="std std-ref">How to optimize convolution using TensorCores</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_tensorcore.py</span></code>)</p></li>
-<li><p><strong>00:01.208</strong>: <a class="reference internal" href="opt_conv_cuda.html#sphx-glr-how-to-optimize-operators-opt-conv-cuda-py"><span class="std std-ref">How to optimize convolution on GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_cuda.py</span></code>)</p></li>
+<li><p><strong>00:32.154</strong>: <a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></li>
+<li><p><strong>00:01.392</strong>: <a class="reference internal" href="opt_conv_tensorcore.html#sphx-glr-how-to-optimize-operators-opt-conv-tensorcore-py"><span class="std std-ref">How to optimize convolution using TensorCores</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_tensorcore.py</span></code>)</p></li>
+<li><p><strong>00:01.199</strong>: <a class="reference internal" href="opt_conv_cuda.html#sphx-glr-how-to-optimize-operators-opt-conv-cuda-py"><span class="std std-ref">How to optimize convolution on GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_cuda.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
index 26fc007..ce73138 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -300,14 +300,14 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autoscheduler-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:01.378</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>04:59.143</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<ul class="simple">
-<li><p><strong>02:26.864</strong>: <a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></li>
-<li><p><strong>01:20.793</strong>: <a class="reference internal" href="tune_network_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-x86-py"><span class="std std-ref">Auto-scheduling a Neural Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_x86.py</span></code>)</p></li>
-<li><p><strong>00:39.472</strong>: <a class="reference internal" href="tune_network_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-cuda-py"><span class="std std-ref">Auto-scheduling a Neural Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_cuda.py</span></code>)</p></li>
-<li><p><strong>00:16.980</strong>: <a class="reference internal" href="tune_sparse_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-sparse-x86-py"><span class="std std-ref">Auto-scheduling Sparse Matrix Multiplication on CPU with Custom Sketch Rule</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_sparse_x86.py</span></code>)</p></li>
-<li><p><strong>00:08.855</strong>: <a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></li>
-<li><p><strong>00:08.413</strong>: <a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></li>
+<li><p><strong>02:25.965</strong>: <a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></li>
+<li><p><strong>01:18.612</strong>: <a class="reference internal" href="tune_network_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-x86-py"><span class="std std-ref">Auto-scheduling a Neural Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_x86.py</span></code>)</p></li>
+<li><p><strong>00:38.791</strong>: <a class="reference internal" href="tune_network_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-cuda-py"><span class="std std-ref">Auto-scheduling a Neural Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_cuda.py</span></code>)</p></li>
+<li><p><strong>00:18.642</strong>: <a class="reference internal" href="tune_sparse_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-sparse-x86-py"><span class="std std-ref">Auto-scheduling Sparse Matrix Multiplication on CPU with Custom Sketch Rule</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_sparse_x86.py</span></code>)</p></li>
+<li><p><strong>00:08.860</strong>: <a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></li>
+<li><p><strong>00:08.274</strong>: <a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
index 2ee26ae..e61c908 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
@@ -469,1106 +469,543 @@ cooperative fetching, unrolling and operator fusion.</p>
data: Buffer(data_2: Pointer(float32), float32, [25088], []),
kernel: Buffer(kernel_2: Pointer(float32), float32, [2359296], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 32;
- allocate(conv2d_nchw: Pointer(local float32), float32, [7]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [1296]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [2304]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [7], [], scope="local", align=16)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [16]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [2016]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [16], [], scope="local", align=64)[0] = 0f32
conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[2] = 0f32
conv2d_nchw_1[3] = 0f32
conv2d_nchw_1[4] = 0f32
conv2d_nchw_1[5] = 0f32
conv2d_nchw_1[6] = 0f32
- for (rc.outer.outer: int32, 0, 32) {
- let cse_var_2: int32 = (rc.outer.outer*784)
- let cse_var_1: int32 = (rc.outer.outer*144)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [1296], [], scope="shared")[(threadIdx.x_1*6)] = @tir.if_then_else(((((3 <= floormod((threadIdx.x_1*2), 27)) && (floormod((threadIdx.x_1*6), 81) < 72)) && (1 <= floormod((threadIdx.x_1*6), 9))) && (floormod((threadIdx.x_1*6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1*2), 27)*49)) + (floordiv(floormod((threadIdx.x_1*2), 27), 3)*7)) + floormod((threadIdx.x_1*6), 9)) - 8 [...]
- pad_temp.shared_1[((threadIdx.x_1*6) + 1)] = @tir.if_then_else(((((3 <= floormod((threadIdx.x_1*2), 27)) && (floormod(((threadIdx.x_1*6) + 1), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 1), 9))) && (floormod(((threadIdx.x_1*6) + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1*2), 27)*49)) + (floordiv(floormod((threadIdx.x_1*2), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 1), 9)) - 8)], 0f32, dtype=float32)
- pad_temp.shared_1[((threadIdx.x_1*6) + 2)] = @tir.if_then_else(((((3 <= floormod((threadIdx.x_1*2), 27)) && (floormod(((threadIdx.x_1*6) + 2), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 2), 9))) && (floormod(((threadIdx.x_1*6) + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1*2), 27)*49)) + (floordiv(floormod((threadIdx.x_1*2), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 2), 9)) - 8)], 0f32, dtype=float32)
- pad_temp.shared_1[((threadIdx.x_1*6) + 3)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 1), 27)) && (floormod(((threadIdx.x_1*6) + 3), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 3), 9))) && (floormod(((threadIdx.x_1*6) + 3), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 1), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 1), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 3), 9)) - 8)], 0f32, dtype=float32)
- pad_temp.shared_1[((threadIdx.x_1*6) + 4)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 1), 27)) && (floormod(((threadIdx.x_1*6) + 4), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 4), 9))) && (floormod(((threadIdx.x_1*6) + 4), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 1), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 1), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 4), 9)) - 8)], 0f32, dtype=float32)
- pad_temp.shared_1[((threadIdx.x_1*6) + 5)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 1), 27)) && (floormod(((threadIdx.x_1*6) + 5), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 5), 9))) && (floormod(((threadIdx.x_1*6) + 5), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 1), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 1), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 5), 9)) - 8)], 0f32, dtype=float32)
- }
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 672)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 224), 27)) && (floormod(((threadIdx.x_1*6) + 24), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 6), 9))) && (floormod(((threadIdx.x_1*6) + 6), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 224), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 224), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 6), 9)) - 8)], 0f32, dty [...]
- }
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 673)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 224), 27)) && (floormod(((threadIdx.x_1*6) + 25), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 7), 9))) && (floormod(((threadIdx.x_1*6) + 7), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 224), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 224), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 7), 9)) - 8)], 0f32, dty [...]
- }
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 674)] = @tir.if_then_else(((((3 <= floormod(((threadIdx.x_1*2) + 224), 27)) && (floormod(((threadIdx.x_1*6) + 26), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 8), 9))) && (floormod(((threadIdx.x_1*6) + 8), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 224), 27)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 224), 27), 3)*7)) + floormod(((threadIdx.x_1*6) + 8), 9)) - 8)], 0f32, dty [...]
+ conv2d_nchw_1[7] = 0f32
+ conv2d_nchw_1[8] = 0f32
+ conv2d_nchw_1[9] = 0f32
+ conv2d_nchw_1[10] = 0f32
+ conv2d_nchw_1[11] = 0f32
+ conv2d_nchw_1[12] = 0f32
+ conv2d_nchw_1[13] = 0f32
+ conv2d_nchw_1[14] = 0f32
+ conv2d_nchw_1[15] = 0f32
+ for (rc.outer.outer: int32, 0, 16) {
+ for (ry.outer.outer: int32, 0, 3) {
+ let cse_var_4: int32 = (rc.outer.outer*1568)
+ let cse_var_3: int32 = (ry.outer.outer*7)
+ let cse_var_2: int32 = (rc.outer.outer*288)
+ let cse_var_1: int32 = (ry.outer.outer*3)
+ {
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [2016], [], scope="shared")[(threadIdx.x_1*12)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1*12), 9))) && (floormod((threadIdx.x_1*12), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod [...]
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 1), 9))) && (floormod(((threadIdx.x_1*12) + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 1), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 2)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*4), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 2), 9))) && (floormod(((threadIdx.x_1*12) + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 2), 9)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 3)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 3), 9))) && (floormod(((threadIdx.x_1*12) + 3), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 1), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 3), 9)) - 8)] [...]
+ pad_temp.shared_1[((threadIdx.x_1*12) + 4)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 4), 9))) && (floormod(((threadIdx.x_1*12) + 4), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 1), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 4), 9)) - 8)] [...]
+ pad_temp.shared_1[((threadIdx.x_1*12) + 5)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 1), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 5), 9))) && (floormod(((threadIdx.x_1*12) + 5), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 1), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 5), 9)) - 8)] [...]
+ pad_temp.shared_1[((threadIdx.x_1*12) + 6)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 6), 9))) && (floormod(((threadIdx.x_1*12) + 6), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 2), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 6), 9)) - 8)] [...]
+ pad_temp.shared_1[((threadIdx.x_1*12) + 7)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 7), 9))) && (floormod(((threadIdx.x_1*12) + 7), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 2), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 7), 9)) - 8)] [...]
+ pad_temp.shared_1[((threadIdx.x_1*12) + 8)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 2), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 8), 9))) && (floormod(((threadIdx.x_1*12) + 8), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 2), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 8), 9)) - 8)] [...]
+ pad_temp.shared_1[((threadIdx.x_1*12) + 9)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7)) < 8)) && (1 <= floormod((threadIdx.x_1*12), 9))) && (floormod((threadIdx.x_1*12), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod((threadIdx.x_1*12), 9)) - 1)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*12) + 10)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 1), 9))) && (floormod(((threadIdx.x_1*12) + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 1), 9)) - 1)], 0f32, [...]
+ pad_temp.shared_1[((threadIdx.x_1*12) + 11)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 2), 9))) && (floormod(((threadIdx.x_1*12) + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 2), 9)) - 1)], 0f32, [...]
}
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 675)] = @tir.if_then_else(((((1 <= floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)) && (floormod(((threadIdx.x_1*6) + 27), 81) < 72)) && (1 <= floormod((threadIdx.x_1*6), 9))) && (floormod((threadIdx.x_1*6), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 225), 27)*49)) + (floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)*7)) + floormod((threadIdx.x_1*6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1176)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 6), 9))) && (floormod(((threadIdx.x_1*12) + 6), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 6) [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1177)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 7), 9))) && (floormod(((threadIdx.x_1*12) + 7), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 7) [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1178)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 392), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 8), 9))) && (floormod(((threadIdx.x_1*12) + 8), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 8) [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1179)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7)) < 8)) && (1 <= floormod((threadIdx.x_1*12), 9))) && (floormod((threadIdx.x_1*12), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod((threadIdx.x_1*12), 9)) + 909)], 0f32, dtype=float32)
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1180)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 1), 9))) && (floormod(((threadIdx.x_1*12) + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 1), 9)) + 909)], [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1181)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*4), 3) + 5), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 2), 9))) && (floormod(((threadIdx.x_1*12) + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1*4), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 2), 9)) + 909)], [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1182)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 3), 9))) && (floormod(((threadIdx.x_1*12) + 3), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 394), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 3) [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1183)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 4), 9))) && (floormod(((threadIdx.x_1*12) + 4), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 394), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 4) [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1184)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*4) + 394), 21), 3) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 5), 9))) && (floormod(((threadIdx.x_1*12) + 5), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 394), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_1*12) + 5) [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1185)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 6), 9))) && (floormod(((threadIdx.x_1*12) + 6), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_ [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1186)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 7), 9))) && (floormod(((threadIdx.x_1*12) + 7), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_ [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 70), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*12) + 1187)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*4) + 392), 3) + 1), 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*12) + 8), 9))) && (floormod(((threadIdx.x_1*12) + 8), 9) < 8)), data[((((cse_var_4 + (floordiv(((threadIdx.x_1*4) + 392), 3)*7)) + cse_var_3) + floormod(((threadIdx.x_ [...]
+ }
}
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 676)] = @tir.if_then_else(((((1 <= floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)) && (floormod(((threadIdx.x_1*6) + 28), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 1), 9))) && (floormod(((threadIdx.x_1*6) + 1), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 225), 27)*49)) + (floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)*7)) + floormod(((threadIdx.x_1*6) + 1), 9)) - 8)], 0f [...]
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope="shared")[threadIdx.x_2] = kernel[((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 96)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 96), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 49), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 98), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 196)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 98), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 196), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 294)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 147), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 2), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 196), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 392), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 490)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 245), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 490), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 588)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 294), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 4), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 686)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 343), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 686), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 392), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 784), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 882)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 441), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 6), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 980)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 490), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 980), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1078)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 539), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1078), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1176)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 588), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1274)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 637), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1274), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1372)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 686), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1372), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1470)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 735), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 10), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 784), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1568), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1666)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 833), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1666), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1764)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 882), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 12), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1862)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 931), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1862), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 1960)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 980), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1960), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2058)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1029), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 14), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2156)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1078), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2156), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2254)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1127), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2254), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2352)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1176), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 16), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2450)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1225), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2450), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2548)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1274), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2548), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2646)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1323), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 18), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2744)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1372), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2744), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2842)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1421), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2842), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ kernel.shared_1[(threadIdx.x_2 + 2940)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1470), 48)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 20), 32)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ if @tir.likely((threadIdx.x_2 < 34), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 3038)] = kernel[((((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 1519), 48)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 3038), 96), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
}
- if @tir.likely((threadIdx.x_1 < 104), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*6) + 677)] = @tir.if_then_else(((((1 <= floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)) && (floormod(((threadIdx.x_1*6) + 29), 81) < 72)) && (1 <= floormod(((threadIdx.x_1*6) + 2), 9))) && (floormod(((threadIdx.x_1*6) + 2), 9) < 8)), data[((((cse_var_2 + (floordiv(((threadIdx.x_1*2) + 225), 27)*49)) + (floormod((floordiv((threadIdx.x_1*2), 3) + 3), 9)*7)) + floormod(((threadIdx.x_1*6) + 2), 9)) - 8)], 0f [...]
+ for (rc.outer.inner: int32, 0, 4) {
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24))]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 96)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 97)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 2)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 98)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 3)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 99)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 4)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 100)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 5)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 101)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 6)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 102)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 7)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 103)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 8)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 104)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 9)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 105)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 10)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 106)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 11)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 107)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 12)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 108)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 13)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 109)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 14)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 110)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 15)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 111)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 16)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 112)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 17)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 113)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 18)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 114)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 19)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 115)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 20)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 116)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 21)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 117)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 22)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 118)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 23)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 119)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 192)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 288)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 193)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 289)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 194)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 290)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 195)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 291)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 196)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 292)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 197)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 293)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 198)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 294)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 199)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 295)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 200)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 296)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 201)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 297)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 202)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 298)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 203)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 299)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 204)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 300)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 205)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 301)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 206)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 302)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 207)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 303)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 208)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 304)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 209)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 305)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 210)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 306)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 211)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 307)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 212)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 308)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 213)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 309)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 214)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 310)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 215)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 311)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 384)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 480)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 385)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 481)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 386)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 482)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 387)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 483)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 388)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 484)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 389)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 485)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 390)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 486)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 391)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 487)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 392)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 488)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 393)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 489)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 394)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 490)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 395)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 491)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 396)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 492)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 397)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 493)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 398)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 494)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 399)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 495)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 400)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 496)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 401)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 497)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 402)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 498)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 403)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 499)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 404)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 500)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 405)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 501)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 406)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 502)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 407)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 503)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 576)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 672)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 577)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 673)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 578)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 674)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 579)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 675)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 580)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 676)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 581)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 677)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 582)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 678)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 583)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 679)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 584)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 680)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 585)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 681)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 586)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 682)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 587)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 683)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 588)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 684)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 589)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 685)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 590)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 686)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 591)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 687)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 592)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 688)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 593)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 689)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 594)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 690)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 595)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 691)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 596)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 692)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 597)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 693)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 598)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 694)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 599)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 695)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 768)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 864)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 769)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 865)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 770)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 866)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 771)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 867)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 772)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 868)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 773)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 869)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 774)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 870)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 775)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 871)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 776)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 872)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 777)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 873)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 778)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 874)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 779)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 875)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 780)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 876)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 781)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 877)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 782)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 878)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 783)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 879)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 784)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 880)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 785)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 881)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 786)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 882)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 787)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 883)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 788)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 884)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 789)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 885)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 790)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 886)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 791)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 887)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 960)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1056)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 961)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1057)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 962)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1058)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 963)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1059)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 964)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1060)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 965)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1061)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 966)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1062)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 967)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1063)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 968)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1064)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 969)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1065)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 970)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1066)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 971)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1067)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 972)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1068)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 973)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1069)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 974)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1070)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 975)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1071)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 976)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1072)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 977)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1073)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 978)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1074)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 979)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1075)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 980)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1076)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 981)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1077)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 982)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1078)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 983)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1079)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1152)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1248)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1153)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1249)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1154)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1250)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1155)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1251)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1156)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1252)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1157)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1253)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1158)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1254)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1159)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1255)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1160)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1256)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1161)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1257)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1162)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1258)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1163)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1259)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1164)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1260)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1165)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1261)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1166)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1262)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1167)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1263)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1168)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1264)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1169)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1265)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1170)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1266)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1171)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1267)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1172)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1268)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1173)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1269)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1174)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1270)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1175)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1271)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1344)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1440)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1345)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1441)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1346)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1442)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1347)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1443)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1348)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1444)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1349)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1445)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1350)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1446)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1351)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1447)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1352)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1448)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1353)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1449)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1354)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1450)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1355)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1451)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1356)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1452)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1357)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1453)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1358)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1454)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1359)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 315)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1455)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1360)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 316)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1456)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1361)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 317)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1457)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1362)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 378)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1458)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1363)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 379)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1459)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1364)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 380)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1460)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1365)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 441)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1461)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1366)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 442)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1462)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1367)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[((((rc.outer.inner*504) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 443)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*24)) + 1463)]))
}
}
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1: Buffer(kernel.shared, float32, [2304], [], scope="shared")[threadIdx.x_2] = kernel[(((blockIdx.x*73728) + cse_var_1) + threadIdx.x_2)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 7), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 112), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 14), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 80), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 21), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 28), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 35), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 128), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 42), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 96), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 49), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 56), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1008)] = kernel[((((blockIdx.x*73728) + cse_var_1) + threadIdx.x_2) + 32256)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 70), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 112), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1232)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 77), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 80), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 84), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1456)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 91), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 98), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 128), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1680)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 105), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 96), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 112), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 1904)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 119), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel[((((blockIdx.x*73728) + cse_var_1) + threadIdx.x_2) + 64512)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 2128)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 133), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 112), 144))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- if @tir.likely((threadIdx.x_2 < 64), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((blockIdx.x*73728) + (floordiv((floordiv(threadIdx.x_2, 16) + 140), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 80), 144))]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 324)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 405)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 486)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 333)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 414)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 495)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 576)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 342)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 423)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 585)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 351)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 432)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 513)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 594)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 360)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 522)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 603)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 369)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 450)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 531)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 612)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*144)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 9)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 18)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 27)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 36)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 459)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 45)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 540)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 54)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 621)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 63)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 325)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 487)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 568)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 334)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 415)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 496)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 577)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 424)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 505)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 586)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 352)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 433)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 514)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 595)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 361)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 523)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 604)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 370)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 451)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 532)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 613)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 1)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 10)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 19)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 28)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 37)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 460)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 46)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 541)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 55)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 622)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 64)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 326)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 407)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 488)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 569)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 335)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 416)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 497)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 578)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 344)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 425)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 506)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 587)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 353)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 515)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 596)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 362)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 524)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 605)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 452)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 533)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 614)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 2)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 11)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 20)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 29)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 38)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 461)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 47)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 542)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 56)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 623)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 65)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 333)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 414)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 495)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 576)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 342)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 423)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 585)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 351)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 432)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 513)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 594)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 360)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 522)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 603)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 369)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 450)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 531)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 612)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 459)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 540)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 621)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 3)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 12)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 21)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 30)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 387)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 39)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 468)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 48)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 549)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 57)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 66)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 334)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 415)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 496)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 577)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 424)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 505)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 586)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 352)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 433)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 514)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 595)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 361)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 523)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 604)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 370)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 451)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 532)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 613)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 460)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 541)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 622)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 4)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 13)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 22)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 31)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 388)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 40)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 49)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 550)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 58)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 631)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 67)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 335)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 416)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 497)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 578)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 344)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 425)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 506)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 587)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 353)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 515)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 596)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 362)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 524)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 605)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 452)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 533)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 614)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 461)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 542)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 623)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 5)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 14)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 23)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 32)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 389)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 41)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 470)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 50)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 551)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 59)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 632)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 68)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 342)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 423)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 585)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 351)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 432)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 513)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 594)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 360)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 522)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 603)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 369)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 450)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 531)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 612)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 459)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 540)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 621)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 387)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 468)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 549)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 6)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 15)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 24)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 33)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 396)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 42)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 477)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 51)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 558)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 60)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 639)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 69)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 424)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 505)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 586)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 352)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 433)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 514)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 595)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 361)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 523)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 604)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 370)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 451)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 532)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 613)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 460)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 541)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 622)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 388)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 550)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 631)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 7)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 16)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 25)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 34)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 397)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 43)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 478)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 52)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 559)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 61)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 640)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 70)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 344)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 425)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 506)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 587)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 353)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 515)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 596)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 362)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 524)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 605)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 452)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 533)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 614)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 461)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 542)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 623)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 389)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 470)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 551)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 632)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 8)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 17)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 26)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 35)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 398)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 44)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 479)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 53)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 560)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 62)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 641)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 71)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 648)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 729)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 810)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 891)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 972)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1053)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1215)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 657)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 738)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 900)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 981)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1062)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1143)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 666)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 747)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 828)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 909)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 990)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1152)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1233)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 675)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 837)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 918)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 999)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1080)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1242)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 684)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 765)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 846)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 927)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1089)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1170)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1251)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 774)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 855)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 936)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1017)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1098)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1179)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 702)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 72)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 783)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 81)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 864)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 90)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 99)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1026)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 108)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1107)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 117)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1188)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 126)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1269)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 135)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 649)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 730)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 811)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 892)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 973)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1054)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 658)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 739)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 820)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 901)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 982)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1063)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 667)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 748)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 829)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 910)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 991)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1072)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 676)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 757)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 838)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 919)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1000)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1081)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 685)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 766)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 847)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 928)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1009)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1090)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 694)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 775)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 856)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 937)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1018)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1099)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 703)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 73)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 784)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 82)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 865)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 91)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 946)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 100)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1027)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 109)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 118)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 127)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 136)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 650)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 731)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 812)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 893)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 974)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1055)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 659)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 740)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 821)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 902)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 983)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1064)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 668)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 749)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 830)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 911)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 992)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1073)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 677)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 758)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 839)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 920)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1001)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1082)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 686)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 767)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 848)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 929)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1010)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1091)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 695)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 776)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 857)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 938)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1019)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 704)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 74)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 785)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 83)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 866)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 92)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 947)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 101)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1028)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 110)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 119)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 128)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 137)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 657)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 738)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 900)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 981)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1062)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1143)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 666)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 747)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 828)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 909)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 990)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1152)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1233)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 675)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 837)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 918)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 999)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1080)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1242)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 684)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 765)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 846)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 927)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1089)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1170)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1251)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 774)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 855)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 936)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1017)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1098)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1179)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 702)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 783)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 864)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1026)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1107)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1188)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1269)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 711)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 75)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 792)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 84)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 873)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 93)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 954)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 102)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1035)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 111)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1116)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 120)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 129)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1278)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 138)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 658)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 739)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 820)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 901)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 982)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1063)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 667)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 748)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 829)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 910)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 991)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1072)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 676)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 757)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 838)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 919)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1000)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1081)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 685)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 766)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 847)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 928)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1009)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1090)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 694)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 775)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 856)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 937)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1018)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1099)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 703)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 784)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 865)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 946)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1027)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 712)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 76)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 793)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 85)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 874)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 94)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 955)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 103)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1036)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 112)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 121)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 130)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 139)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 659)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 740)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 821)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 902)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 983)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1064)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 668)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 749)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 830)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 911)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 992)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1073)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 677)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 758)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 839)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 920)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1001)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1082)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 686)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 767)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 848)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 929)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1010)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1091)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 695)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 776)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 857)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 938)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1019)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 704)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 785)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 866)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 947)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1028)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 713)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 77)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 794)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 86)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 875)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 95)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 956)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 104)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1037)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 113)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 122)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 131)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 140)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 666)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 747)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 828)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 909)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 990)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1152)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1233)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 675)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 837)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 918)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 999)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1080)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1242)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 684)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 765)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 846)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 927)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1089)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1170)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1251)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 774)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 855)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 936)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1017)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1098)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1179)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 702)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 783)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 864)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1026)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1107)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1188)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1269)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 711)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 792)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 873)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 954)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1035)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1116)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1278)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 720)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 78)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 801)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 87)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 96)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 963)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 105)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1044)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 114)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1125)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 123)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1206)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 132)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 141)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 667)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 748)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 829)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 910)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 991)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1072)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 676)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 757)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 838)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 919)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1000)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1081)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 685)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 766)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 847)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 928)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1009)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1090)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 694)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 775)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 856)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 937)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1018)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1099)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 703)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 784)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 865)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 946)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1027)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 712)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 793)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 874)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 955)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1036)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 721)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 79)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 802)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 88)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 883)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 97)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 964)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 106)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1045)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 115)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 124)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 133)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 142)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 668)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 749)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 830)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 911)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 992)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1073)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 677)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 758)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 839)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 920)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1001)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1082)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 686)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 767)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 848)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 929)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1010)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1091)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 695)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 776)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 857)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 938)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1019)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 704)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 785)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 866)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 947)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1028)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 713)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 794)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 875)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 956)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1037)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 722)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 80)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 803)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 89)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 884)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 98)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 965)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 107)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1046)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 116)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 125)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 134)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + 143)]))
}
}
- for (i2.inner: int32, 0, 7) {
- compute[((((blockIdx.x*784) + (floordiv(threadIdx.x, 7)*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[i2.inner] + bias[((blockIdx.x*16) + floordiv(threadIdx.x, 7))]), 0f32)
+ for (i1.inner: int32, 0, 16) {
+ compute[((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*784)) + (i1.inner*49)) + floormod(threadIdx.x, 49))] = max((conv2d_nchw_1[i1.inner] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*16)) + i1.inner)]), 0f32)
}
}
}
@@ -1606,7 +1043,7 @@ cooperative fetching, unrolling and operator fusion.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.155 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.274 ms
</pre></div>
</div>
</div>
@@ -1636,33 +1073,33 @@ conv2d_nchw_nn_o_i, conv2d_nchw_nn_i = s[conv2d_nchw].split(conv2d_nchw_nn, fact
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
-conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
-conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
-conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=16)
+conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
+conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=8)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=2)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
-conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=7)
-conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
+conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
+conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=8)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
-conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
-conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
-conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
+conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
+conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
+conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=16)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=16)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=2)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
-compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=7)
-compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
+compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
@@ -1685,12 +1122,12 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=6)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=12)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 1024)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
@@ -1710,10 +1147,10 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(112) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[7];
- __shared__ float pad_temp_shared[1296];
- __shared__ float kernel_shared[2304];
+extern "C" __global__ void __launch_bounds__(98) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[16];
+ __shared__ float pad_temp_shared[2016];
+ __shared__ float kernel_shared[3072];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
@@ -1721,1067 +1158,491 @@ extern "C" __global__ void __launch_bounds__(112) default_function_ker
conv2d_nchw[4] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 32; ++rc_outer_outer) {
- __syncthreads();
- pad_temp_shared[(((int)threadIdx.x) * 6)] = (((((3 <= ((((int)threadIdx.x) * 2) % 27)) && (((((int)threadIdx.x) * 6) % 81) < 72)) && (1 <= ((((int)threadIdx.x) * 6) % 9))) && (((((int)threadIdx.x) * 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 2) / 27) * 49)) + ((((((int)threadIdx.x) * 2) % 27) / 3) * 7)) + ((((int)threadIdx.x) * 6) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 1)] = (((((3 <= ((((int)threadIdx.x) * 2) % 27)) && ((((((int)threadIdx.x) * 6) + 1) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 1) % 9))) && ((((((int)threadIdx.x) * 6) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 2) / 27) * 49)) + ((((((int)threadIdx.x) * 2) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 1) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 2)] = (((((3 <= ((((int)threadIdx.x) * 2) % 27)) && ((((((int)threadIdx.x) * 6) + 2) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 2) % 9))) && ((((((int)threadIdx.x) * 6) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 2) / 27) * 49)) + ((((((int)threadIdx.x) * 2) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 2) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 3)] = (((((3 <= (((((int)threadIdx.x) * 2) + 1) % 27)) && ((((((int)threadIdx.x) * 6) + 3) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 3) % 9))) && ((((((int)threadIdx.x) * 6) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 1) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 3) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 4)] = (((((3 <= (((((int)threadIdx.x) * 2) + 1) % 27)) && ((((((int)threadIdx.x) * 6) + 4) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 4) % 9))) && ((((((int)threadIdx.x) * 6) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 1) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 4) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[((((int)threadIdx.x) * 6) + 5)] = (((((3 <= (((((int)threadIdx.x) * 2) + 1) % 27)) && ((((((int)threadIdx.x) * 6) + 5) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 5) % 9))) && ((((((int)threadIdx.x) * 6) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 1) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 5) % 9)) - 8)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 672)] = (((((3 <= (((((int)threadIdx.x) * 2) + 8) % 27)) && ((((((int)threadIdx.x) * 6) + 24) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 6) % 9))) && ((((((int)threadIdx.x) * 6) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 224) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 8) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 6) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 673)] = (((((3 <= (((((int)threadIdx.x) * 2) + 8) % 27)) && ((((((int)threadIdx.x) * 6) + 25) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 7) % 9))) && ((((((int)threadIdx.x) * 6) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 224) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 8) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 7) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 674)] = (((((3 <= (((((int)threadIdx.x) * 2) + 8) % 27)) && ((((((int)threadIdx.x) * 6) + 26) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 8) % 9))) && ((((((int)threadIdx.x) * 6) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 224) / 27) * 49)) + (((((((int)threadIdx.x) * 2) + 8) % 27) / 3) * 7)) + (((((int)threadIdx.x) * 6) + 8) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 675)] = (((((1 <= ((((((int)threadIdx.x) * 2) / 3) + 3) % 9)) && ((((((int)threadIdx.x) * 6) + 27) % 81) < 72)) && (1 <= ((((int)threadIdx.x) * 6) % 9))) && (((((int)threadIdx.x) * 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 225) / 27) * 49)) + (((((((int)threadIdx.x) * 2) / 3) + 3) % 9) * 7)) + ((((int)threadIdx.x) * 6) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 676)] = (((((1 <= ((((((int)threadIdx.x) * 2) / 3) + 3) % 9)) && ((((((int)threadIdx.x) * 6) + 28) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 1) % 9))) && ((((((int)threadIdx.x) * 6) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 225) / 27) * 49)) + (((((((int)threadIdx.x) * 2) / 3) + 3) % 9) * 7)) + (((((int)threadIdx.x) * 6) + 1) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 104) {
- pad_temp_shared[((((int)threadIdx.x) * 6) + 677)] = (((((1 <= ((((((int)threadIdx.x) * 2) / 3) + 3) % 9)) && ((((((int)threadIdx.x) * 6) + 29) % 81) < 72)) && (1 <= (((((int)threadIdx.x) * 6) + 2) % 9))) && ((((((int)threadIdx.x) * 6) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 2) + 225) / 27) * 49)) + (((((((int)threadIdx.x) * 2) / 3) + 3) % 9) * 7)) + (((((int)threadIdx.x) * 6) + 2) % 9)) - 8)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 73728) + (rc_outer_outer * 144)) + ((int)threadIdx.x))];
- kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 112) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 112) % 144))];
- kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 224) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 80) % 144))];
- kernel_shared[(((int)threadIdx.x) + 336)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 336) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 48) % 144))];
- kernel_shared[(((int)threadIdx.x) + 448)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 448) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 16))];
- kernel_shared[(((int)threadIdx.x) + 560)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 560) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 128) % 144))];
- kernel_shared[(((int)threadIdx.x) + 672)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 672) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 96) % 144))];
- kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 784) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 64) % 144))];
- kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 896) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 32))];
- kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 32256)];
- kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1120) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 112) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1232) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 80) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1344) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 48) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1456) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 16))];
- kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1568) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 128) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1680) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 96) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1792) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 64) % 144))];
- kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1904) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 32))];
- kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 64512)];
- kernel_shared[(((int)threadIdx.x) + 2128)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2128) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 112) % 144))];
- if (((int)threadIdx.x) < 64) {
- kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2240) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) + 80))];
+ conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
+ for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[(((int)threadIdx.x) * 12)] = (((((1 <= ((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) * 12) % 9))) && (((((int)threadIdx.x) * 12) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) * 12) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1)] = (((((1 <= ((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 1) % 9))) && ((((((int)threadIdx.x) * 12) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 2)] = (((((1 <= ((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) * 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 2) % 9))) && ((((((int)threadIdx.x) * 12) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 3)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 3) % 9))) && ((((((int)threadIdx.x) * 12) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 3) % 9)) - 8)] : 0 [...]
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 4)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 4) % 9))) && ((((((int)threadIdx.x) * 12) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 4) % 9)) - 8)] : 0 [...]
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 5)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 5) % 9))) && ((((((int)threadIdx.x) * 12) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 1) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 5) % 9)) - 8)] : 0 [...]
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 6)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 6) % 9))) && ((((((int)threadIdx.x) * 12) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 6) % 9)) - 8)] : 0 [...]
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 7)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 7) % 9))) && ((((((int)threadIdx.x) * 12) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 7) % 9)) - 8)] : 0 [...]
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 8)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 8) % 9))) && ((((((int)threadIdx.x) * 12) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 2) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 8) % 9)) - 8)] : 0 [...]
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 9)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 12) % 9))) && (((((int)threadIdx.x) * 12) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) * 12) % 9)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 10)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 1) % 9))) && ((((((int)threadIdx.x) * 12) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 1) % 9)) - 1)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 11)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 2) % 9))) && ((((((int)threadIdx.x) * 12) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 2) % 9)) - 1)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1176)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 6) % 9))) && ((((((int)threadIdx.x) * 12) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 6) % 9)) [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1177)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 7) % 9))) && ((((((int)threadIdx.x) * 12) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 7) % 9)) [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1178)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 8) % 9))) && ((((((int)threadIdx.x) * 12) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 8) % 9)) [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1179)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 12) % 9))) && (((((int)threadIdx.x) * 12) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) * 12) % 9)) + 909)] : 0.000000e+00f);
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1180)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 1) % 9))) && ((((((int)threadIdx.x) * 12) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 1) % 9)) + 909)] : 0. [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1181)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 4) / 3) + 5) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 2) % 9))) && ((((((int)threadIdx.x) * 12) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 2) % 9)) + 909)] : 0. [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1182)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 3) % 9))) && ((((((int)threadIdx.x) * 12) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 394) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 3) % 9)) [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1183)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 4) % 9))) && ((((((int)threadIdx.x) * 12) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 394) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 4) % 9)) [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1184)] = (((((1 <= (((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 4) + 16) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 5) % 9))) && ((((((int)threadIdx.x) * 12) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 394) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) + 5) % 9)) [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1185)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 6) % 9))) && ((((((int)threadIdx.x) * 12) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1186)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 7) % 9))) && ((((((int)threadIdx.x) * 12) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) [...]
+ }
+ if (((int)threadIdx.x) < 70) {
+ pad_temp_shared[((((int)threadIdx.x) * 12) + 1187)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 4) + 392) / 3) + 1) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 12) + 8) % 9))) && ((((((int)threadIdx.x) * 12) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 4) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((((int)threadIdx.x) * 12) [...]
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 98)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 98) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 2) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 196) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 4) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 294)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 294) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 2) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 8) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 490)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 490) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 10) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 588)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 588) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 4) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 686)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 686) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 14) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 16) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 882)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 882) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 6) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 980)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 980) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 20) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1078)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1078) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 22) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1176) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 8) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1274)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1274) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 26) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1372) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 28) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1470)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1470) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 10) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1568) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 32) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1666)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1666) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 34) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1764)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1764) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 12) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1862)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1862) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 38) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1960) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 40) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2058)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2058) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 14) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2156)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2156) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 44) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2254)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2254) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 46) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2352) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2450)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2450) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 50) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2548)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2548) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 52) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2646)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2646) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 18) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2744) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2842)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2842) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 58) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2940)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2940) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 20) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ if (((int)threadIdx.x) < 34) {
+ kernel_shared[(((int)threadIdx.x) + 3038)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3038) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 62) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ }
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 4; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24))]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 96)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 97)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 2)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 98)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 3)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 99)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 4)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 100)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 5)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 101)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 6)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 102)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 7)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 103)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 8)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 104)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 9)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 105)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 10)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 106)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 11)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 107)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 12)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 108)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 13)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 109)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 14)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 110)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 15)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 111)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 16)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 112)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 17)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 113)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 18)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 114)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 19)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 115)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 20)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 116)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 21)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 117)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 22)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 118)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 23)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 119)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 192)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 288)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 193)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 289)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 194)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 290)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 195)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 291)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 196)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 292)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 197)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 293)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 198)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 294)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 199)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 295)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 200)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 296)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 201)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 297)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 202)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 298)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 203)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 299)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 204)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 300)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 205)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 301)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 206)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 302)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 207)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 303)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 208)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 304)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 209)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 305)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 210)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 306)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 211)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 307)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 212)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 308)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 213)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 309)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 214)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 310)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 215)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 311)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 384)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 480)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 385)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 481)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 386)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 482)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 387)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 483)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 388)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 484)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 389)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 485)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 390)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 486)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 391)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 487)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 392)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 488)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 393)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 489)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 394)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 490)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 395)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 491)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 396)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 492)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 397)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 493)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 398)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 494)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 399)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 495)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 400)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 496)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 401)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 497)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 402)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 498)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 403)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 499)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 404)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 500)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 405)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 501)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 406)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 502)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 407)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 503)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 576)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 672)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 577)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 673)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 578)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 674)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 579)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 675)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 580)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 676)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 581)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 677)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 582)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 678)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 583)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 679)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 584)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 680)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 585)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 681)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 586)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 682)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 587)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 683)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 588)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 684)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 589)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 685)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 590)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 686)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 591)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 687)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 592)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 688)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 593)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 689)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 594)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 690)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 595)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 691)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 596)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 692)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 597)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 693)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 598)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 694)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 599)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 695)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 768)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 864)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 769)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 865)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 770)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 866)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 771)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 867)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 772)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 868)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 773)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 869)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 774)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 870)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 775)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 871)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 776)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 872)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 777)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 873)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 778)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 874)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 779)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 875)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 780)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 876)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 781)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 877)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 782)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 878)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 783)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 879)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 784)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 880)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 785)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 881)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 786)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 882)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 787)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 883)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 788)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 884)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 789)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 885)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 790)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 886)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 791)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 887)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 960)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1056)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 961)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1057)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 962)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1058)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 963)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1059)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 964)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1060)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 965)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1061)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 966)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1062)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 967)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1063)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 968)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1064)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 969)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1065)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 970)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1066)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 971)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1067)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 972)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1068)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 973)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1069)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 974)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1070)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 975)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1071)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 976)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1072)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 977)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1073)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 978)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1074)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 979)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1075)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 980)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1076)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 981)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1077)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 982)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1078)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 983)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1079)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1152)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1248)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1153)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1249)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1154)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1250)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1155)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1251)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1156)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1252)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1157)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1253)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1158)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1254)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1159)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1255)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1160)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1256)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1161)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1257)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1162)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1258)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1163)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1259)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1164)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1260)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1165)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1261)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1166)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1262)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1167)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1263)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1168)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1264)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1169)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1265)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1170)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1266)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1171)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1267)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1172)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1268)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1173)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1269)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1174)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1270)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1175)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1271)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1344)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1440)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1345)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1441)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1346)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1442)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1347)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1443)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1348)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1444)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1349)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1445)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1350)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1446)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1351)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1447)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1352)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1448)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1353)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1449)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1354)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1450)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1355)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1451)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1356)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1452)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1357)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1453)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1358)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1454)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1359)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 315)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1455)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1360)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 316)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1456)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1361)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 317)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1457)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1362)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 378)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1458)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1363)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 379)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1459)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1364)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 380)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1460)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1365)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 441)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1461)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1366)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 442)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1462)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1367)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 504) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 443)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 24)) + 1463)]));
+ }
}
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 324)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 405)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 486)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 333)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 414)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 495)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 576)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 342)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 423)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 585)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 351)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 432)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 513)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 594)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 360)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 522)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 603)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 369)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 450)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 531)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 612)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[((((int)threadIdx.x) / 7) * 144)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 27)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 36)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 459)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 45)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 540)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 54)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 621)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 63)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 325)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 487)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 334)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 415)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 496)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 577)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 424)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 586)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 352)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 433)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 514)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 595)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 361)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 523)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 604)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 370)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 451)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 532)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 613)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 28)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 37)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 460)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 46)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 541)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 55)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 622)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 64)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 326)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 407)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 488)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 335)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 416)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 497)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 578)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 344)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 425)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 587)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 353)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 434)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 515)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 596)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 362)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 524)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 605)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 371)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 452)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 533)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 614)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 29)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 38)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 461)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 47)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 542)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 56)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 623)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 65)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 333)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 414)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 495)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 576)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 342)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 423)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 585)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 351)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 432)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 513)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 594)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 360)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 522)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 603)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 369)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 450)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 531)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 612)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 459)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 540)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 621)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 30)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 387)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 39)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 468)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 549)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 57)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 66)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 334)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 415)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 496)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 577)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 424)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 586)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 352)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 433)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 514)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 595)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 361)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 523)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 604)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 370)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 451)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 532)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 613)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 460)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 541)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 622)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 31)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 388)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 40)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 49)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 550)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 58)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 67)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 335)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 416)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 497)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 578)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 344)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 425)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 587)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 353)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 434)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 515)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 596)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 362)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 524)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 605)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 371)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 452)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 533)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 614)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 461)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 542)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 623)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 32)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 389)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 41)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 470)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 50)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 551)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 59)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 68)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 342)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 423)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 585)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 351)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 432)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 513)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 594)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 360)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 522)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 603)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 369)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 450)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 531)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 612)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 459)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 540)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 621)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 387)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 468)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 549)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 24)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 33)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 396)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 42)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 477)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 51)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 558)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 60)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 639)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 69)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 424)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 586)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 352)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 433)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 514)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 595)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 361)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 523)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 604)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 370)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 451)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 532)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 613)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 460)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 541)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 622)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 388)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 550)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 25)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 34)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 397)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 43)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 478)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 52)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 559)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 61)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 640)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 70)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 344)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 425)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 587)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 353)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 434)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 515)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 596)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 362)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 524)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 605)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 371)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 452)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 533)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 614)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 461)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 542)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 623)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 389)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 470)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 551)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 26)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 35)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 398)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 44)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 479)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 53)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 560)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 62)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 641)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 71)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 648)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 729)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 810)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 891)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 972)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1053)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1215)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 657)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 738)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 819)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 900)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 981)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1062)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1143)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 666)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 747)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 828)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 909)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 990)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1152)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 675)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 756)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 837)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 918)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 999)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1080)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1242)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 684)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 765)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 846)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 927)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1089)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1251)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 693)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 774)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 855)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 936)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1017)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1098)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1179)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 702)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 72)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 783)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 81)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 864)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 90)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 945)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 99)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1026)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 108)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 117)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1188)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 126)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1269)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 135)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 649)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 730)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 811)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 892)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 973)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1054)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 658)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 739)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 820)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 901)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 982)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1063)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 667)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 748)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 829)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 910)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 991)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1072)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 676)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 757)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 838)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 919)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1000)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1081)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 685)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 766)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 847)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 928)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1009)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1090)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 694)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 775)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 856)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 937)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1018)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1099)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 703)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 73)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 784)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 82)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 865)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 91)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 946)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 100)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1027)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 109)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 118)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 127)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 136)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 650)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 731)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 812)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 893)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 974)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1055)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 659)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 740)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 821)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 902)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 983)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1064)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 668)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 749)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 830)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 911)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 992)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1073)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 677)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 758)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 839)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 920)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1001)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1082)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 686)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 767)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 848)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 929)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1010)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1091)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 695)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 776)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 857)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 938)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1019)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 704)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 74)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 785)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 83)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 866)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 92)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 947)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 101)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1028)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 110)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 119)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 128)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 137)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 657)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 738)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 819)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 900)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 981)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1062)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1143)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 666)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 747)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 828)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 909)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 990)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1152)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 675)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 756)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 837)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 918)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 999)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1080)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1242)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 684)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 765)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 846)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 927)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1089)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1251)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 693)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 774)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 855)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 936)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1017)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1098)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1179)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 702)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 783)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 864)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 945)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1026)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1188)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1269)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 711)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 75)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 792)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 84)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 873)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 93)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 954)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 102)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1035)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 111)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1116)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 120)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 129)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1278)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 138)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 658)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 739)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 820)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 901)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 982)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1063)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 667)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 748)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 829)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 910)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 991)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1072)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 676)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 757)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 838)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 919)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1000)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1081)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 685)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 766)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 847)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 928)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1009)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1090)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 694)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 775)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 856)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 937)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1018)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1099)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 703)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 784)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 865)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 946)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1027)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 712)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 76)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 793)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 85)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 874)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 94)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 955)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 103)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1036)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 112)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 121)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 130)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 139)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 659)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 740)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 821)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 902)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 983)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1064)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 668)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 749)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 830)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 911)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 992)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1073)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 677)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 758)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 839)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 920)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1001)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1082)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 686)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 767)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 848)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 929)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1010)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1091)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 695)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 776)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 857)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 938)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1019)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 704)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 785)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 866)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 947)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1028)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 713)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 77)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 794)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 86)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 875)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 95)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 956)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 104)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1037)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 113)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 122)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 131)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 140)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 666)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 747)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 828)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 909)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 990)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1152)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 675)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 756)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 837)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 918)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 999)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1080)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1242)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 684)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 765)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 846)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 927)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1089)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1251)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 693)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 774)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 855)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 936)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1017)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1098)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1179)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 702)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 783)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 864)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 945)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1026)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1188)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1269)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 711)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 792)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 873)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 954)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1035)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1116)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1278)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 720)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 78)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 801)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 87)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 882)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 96)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 963)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 105)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1044)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 114)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1125)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 123)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1206)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 132)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 141)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 667)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 748)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 829)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 910)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 991)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1072)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 676)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 757)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 838)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 919)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1000)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1081)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 685)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 766)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 847)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 928)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1009)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1090)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 694)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 775)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 856)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 937)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1018)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1099)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 703)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 784)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 865)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 946)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1027)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 712)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 793)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 874)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 955)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1036)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 721)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 79)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 802)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 88)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 883)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 97)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 964)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 106)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1045)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 115)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 124)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 133)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 142)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 668)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 749)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 830)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 911)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 992)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1073)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 677)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 758)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 839)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 920)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1001)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1082)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 686)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 767)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 848)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 929)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1010)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1091)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 695)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 776)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 857)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 938)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1019)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 704)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 785)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 866)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 947)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1028)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 713)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 794)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 875)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 956)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1037)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 722)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 80)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 803)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 89)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 884)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 98)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 965)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 107)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1046)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 116)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 125)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 134)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + 143)]));
}
- for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
- compute[((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 7) * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[i2_inner] + bias[((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 16; ++i1_inner) {
+ compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 784)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 16)) + i1_inner)]), 0.000000e+00f);
}
}
</pre></div>
@@ -2819,7 +1680,7 @@ In the example below we resume the status and do more 5 trials.</p>
Get devices for measurement successfully!
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 26.864 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 25.965 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/e3e540f3b477c0c52d8eb73e674e8ffd/tune_conv2d_layer_cuda.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_conv2d_layer_cuda.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
index 2418510..5acdaed 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -876,7 +876,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 9.8594 9.8517 9.9018 9.8247 0.0320
+ 9.9032 9.9073 9.9324 9.8701 0.0256
</pre></div>
</div>
</div>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
index 8fb4986..7a0dc8a 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -895,7 +895,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 760.1292 760.3889 760.6901 759.3086 0.5931
+ 748.3389 746.1669 752.9568 745.8929 3.2673
</pre></div>
</div>
</div>
@@ -917,7 +917,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 20.793 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 18.612 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-network-x86-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/e416b94ca1090b0897c0f6e0df95b911/tune_network_x86.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_network_x86.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
index 5fa4af1..9942f13 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -593,36 +593,33 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Lowered TIR:
@main = primfn(placeholder_5: handle, placeholder_6: handle, placeholder_7: handle, placeholder_8: handle, placeholder_9: handle, compute_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {placeholder_4: Buffer(placeholder_10: Pointer(float32), float32, [32768], []),
- placeholder_2: Buffer(placeholder_11: Pointer(float32), float32, [78656], []),
- placeholder_3: Buffer(placeholder_12: Pointer(int32), int32, [4916], []),
+ buffers = {placeholder_4: Buffer(placeholder_10: Pointer(int32), int32, [33], []),
+ placeholder_2: Buffer(placeholder_11: Pointer(int32), int32, [4916], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], []),
- placeholder: Buffer(placeholder_13: Pointer(float32), float32, [65536], []),
- placeholder_1: Buffer(placeholder_14: Pointer(int32), int32, [33], [])}
- buffer_map = {placeholder_9: placeholder, placeholder_8: placeholder_1, compute_1: compute, placeholder_6: placeholder_2, placeholder_7: placeholder_3, placeholder_5: placeholder_4} {
- allocate(compute_3: Pointer(global float32), float32, [4096]), storage_scope = global;
- for (i1.outer: int32, 0, 16) {
- for (i.outer.inner: int32, 0, 16) {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 8) {
- for (j.init: int32, 0, 16) {
- compute_4: Buffer(compute_3, float32, [4096], [])[((((i.outer.inner*256) + (i.inner.init*32)) + (nb_j.inner*16)) + j.init)] = 0f32
- }
+ placeholder: Buffer(placeholder_12: Pointer(float32), float32, [32768], []),
+ placeholder_3: Buffer(placeholder_13: Pointer(float32), float32, [65536], []),
+ placeholder_1: Buffer(placeholder_14: Pointer(float32), float32, [78656], [])}
+ buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, compute_1: compute, placeholder_7: placeholder_2, placeholder_9: placeholder_3, placeholder_8: placeholder_4} {
+ for (i0.outer.i1.outer.fused: int32, 0, 256) "parallel" {
+ allocate(compute_3: Pointer(global float32), float32, [256]), storage_scope = global {
+ for (i.inner.init: int32, 0, 16) {
+ for (j.init: int32, 0, 16) {
+ compute_4: Buffer(compute_3, float32, [256], [])[((i.inner.init*16) + j.init)] = 0f32
}
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((i1.outer*2) + nb_j.inner) in (placeholder_1[(cse_var_1 + 1)] - placeholder_1[cse_var_1])) {
- for (i.inner: int32, 0, 8) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((i1.outer*2) + nb_j.inner)
- let cse_var_2: int32 = ((((i.outer.inner*256) + (i.inner*32)) + (nb_j.inner*16)) + j)
- compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_2[(((placeholder_1[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder_4[(((i.outer.inner*2048) + (i.inner*256)) + placeholder_3[(placeholder_1[cse_var_3] + elem_idx)])], 0f32)))
- }
+ }
+ for (elem_idx: int32, 0, let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_4[(cse_var_1 + 1)] - placeholder_4[cse_var_1])) {
+ for (i.inner: int32, 0, 16) {
+ for (j: int32, 0, 16) {
+ let cse_var_3: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ let cse_var_2: int32 = ((i.inner*16) + j)
+ compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_1[(((placeholder_4[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + (i.inner*256)) + placeholder_2[(placeholder_4[cse_var_3] + elem_idx)])], 0f32)))
}
}
}
- }
- for (i0.inner: int32, 0, 128) {
- let cse_var_4: int32 = ((i0.inner*512) + (i1.outer*32))
- compute[ramp(cse_var_4, 1, 32)] = max((compute_4[ramp((i0.inner*32), 1, 32)] + placeholder[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 16) {
+ let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*8192) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
+ compute[ramp(cse_var_4, 1, 16)] = max((compute_4[ramp((i0.inner*16), 1, 16)] + placeholder_3[ramp(cse_var_4, 1, 16)]), broadcast(0f32, 16))
+ }
}
}
}
@@ -660,7 +657,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.411 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.538 ms
</pre></div>
</div>
<div class="admonition note">
diff --git a/docs/how_to/tune_with_autotvm/sg_execution_times.html b/docs/how_to/tune_with_autotvm/sg_execution_times.html
index 9208f1e..32c3689 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -300,13 +300,13 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:44.397</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:43.813</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:43.520</strong>: <a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></li>
-<li><p><strong>00:00.230</strong>: <a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></li>
-<li><p><strong>00:00.217</strong>: <a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></li>
-<li><p><strong>00:00.216</strong>: <a class="reference internal" href="tune_relay_arm.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-arm-py"><span class="std std-ref">Auto-tuning a Convolutional Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_arm.py</span></code>)</p></li>
-<li><p><strong>00:00.215</strong>: <a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></li>
+<li><p><strong>00:43.071</strong>: <a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></li>
+<li><p><strong>00:00.196</strong>: <a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></li>
+<li><p><strong>00:00.184</strong>: <a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></li>
+<li><p><strong>00:00.183</strong>: <a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></li>
+<li><p><strong>00:00.180</strong>: <a class="reference internal" href="tune_relay_arm.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-arm-py"><span class="std std-ref">Auto-tuning a Convolutional Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_arm.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
index cfbd8a4..d4dae27 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -1142,8 +1142,8 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2885496
-No: 6 GFLOPS: 100.79/100.79 result: MeasureResult(costs=(0.0022969085208333333,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5907156467437744, timestamp=1647466290.0848103) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
-No: 7 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 6 GFLOPS: 94.20/94.20 result: MeasureResult(costs=(0.0024575866875,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5868585109710693, timestamp=1647477956.84365) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
+No: 7 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1266,7 +1266,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6225319
-No: 8 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 8 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1389,7 +1389,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,943546
-No: 9 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 9 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1512,7 +1512,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2868708
-No: 10 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 10 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
res = future.result()
File "/usr/lib/python3.6/concurrent/futures/_base.py", line 432, in result
@@ -1530,7 +1530,7 @@ No: 10 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
TimeoutError
[('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4691833
-No: 11 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 11 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1653,7 +1653,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1042124
-No: 12 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 12 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1776,7 +1776,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10013405
-No: 13 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 13 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1899,7 +1899,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6732082
-No: 14 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 14 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2022,7 +2022,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7536735
-No: 15 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 15 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2145,7 +2145,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,482121
-No: 16 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 16 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2268,7 +2268,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2824525
-No: 17 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 17 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2391,7 +2391,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4559286
-No: 18 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 18 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2514,7 +2514,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9677544
-No: 19 GFLOPS: 0.00/100.79 result: Traceback (most recent call last):
+No: 19 GFLOPS: 0.00/94.20 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 721, in __call__
yield remote, remote.load_module(os.path.split(build_result.filename)[1])
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 685, in run_through_rpc
@@ -2610,7 +2610,7 @@ tvm._ffi.base.TVMError: Traceback (most recent call last):
15: _PyEval_EvalFrameDefault
14: 0x000000000050a7f2
13: _PyObject_FastCallKeywords
- 12: 0x00007f1aafd34c12
+ 12: 0x00007f5502403c12
11: _ctypes_callproc
10: ffi_call
9: ffi_call_unix64
@@ -2680,7 +2680,7 @@ Traceback (most recent call last):
24: 0x000000000050a6bc
23: 0x0000000000509988
22: _PyEval_EvalFrameDe [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6390073
-No: 20 GFLOPS: 144.43/144.43 result: MeasureResult(costs=(0.00160282077,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4138598442077637, timestamp=1647466315.689689) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
+No: 20 GFLOPS: 144.18/144.18 result: MeasureResult(costs=(0.0016056626900000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3832473754882812, timestamp=1647477981.520465) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2719,7 +2719,7 @@ and measure running time.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Best config:
[('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
-Time cost of this operator: 0.001982
+Time cost of this operator: 0.001978
</pre></div>
</div>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autotvm-tune-conv2d-cuda-py">
diff --git a/docs/how_to/work_with_microtvm/micro_autotune.html b/docs/how_to/work_with_microtvm/micro_autotune.html
index a76a493..9fe4b3b 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -553,10 +553,10 @@ the tuned operator.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 313.0 98.753 (1, 2, 10, 10, 3) 2 1
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.0 0.947 (1, 6, 10, 10) 1 1
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.953 0.301 (1, 1, 10, 10, 3) 1 1
-Total_time - 316.953 - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 310.7 98.704 (1, 2, 10, 10, 3) 2 1
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.144 0.999 (1, 6, 10, 10) 1 1
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.934 0.297 (1, 1, 10, 10, 3) 1 1
+Total_time - 314.778 - - - -
</pre></div>
</div>
</div>
@@ -608,10 +608,10 @@ Total_time -
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 92.6 97.204 (1, 6, 10, 10, 1) 2 1
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.741 1.828 (1, 6, 10, 10) 1 1
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.922 0.968 (1, 1, 10, 10, 3) 1 1
-Total_time - 95.263 - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 80.35 96.752 (1, 6, 10, 10, 1) 2 1
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.77 2.132 (1, 6, 10, 10) 1 1
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.927 1.116 (1, 1, 10, 10, 3) 1 1
+Total_time - 83.047 - - - -
</pre></div>
</div>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
diff --git a/docs/how_to/work_with_microtvm/micro_tflite.html b/docs/how_to/work_with_microtvm/micro_tflite.html
index cb512ac..2e4ce02 100644
--- a/docs/how_to/work_with_microtvm/micro_tflite.html
+++ b/docs/how_to/work_with_microtvm/micro_tflite.html
@@ -614,7 +614,7 @@ QEMU VM based on BOARD. In the example below the x86 arch is selected and a x86
#ifdef __cplusplus
extern "C"
#endif
-TVM_DLL int32_t tvmgen_default_fused_nn_dense_add(void* args, int32_t* arg_type_ids, int32_t num_args, void* out_ret_value, int32_t* out_ret_tcode, void* resource_handle) {
+TVM_DLL int32_t tvmgen_default_fused_reshape_1(void* args, int32_t* arg_type_ids, int32_t num_args, void* out_ret_value, int32_t* out_ret_tcode, void* resource_handle) {
void* arg_placeholder = (((TVMValue*)args)[0].v_handle);
- .
- ./codegen
diff --git a/docs/how_to/work_with_microtvm/micro_tvmc.html b/docs/how_to/work_with_microtvm/micro_tvmc.html
index 9c59e5b..93950c8 100644
--- a/docs/how_to/work_with_microtvm/micro_tvmc.html
+++ b/docs/how_to/work_with_microtvm/micro_tvmc.html
@@ -398,15 +398,15 @@ containing a file for each piece of the TVM compiler output which can be used on
TVM environment. Read more about <a class="reference external" href="https://tvm.apache.org/docs//arch/model_library_format.html">Model Library Format</a>.</p>
<p>Here, we generate a MLF file for <code class="docutils literal notranslate"><span class="pre">qemu_x86</span></code> Zephyr board. To generate MLF output for the <code class="docutils literal notranslate"><span class="pre">magic_wand</span></code> tflite model:</p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>tvmc compile magic_wand.tflite <span class="se">\</span>
- --target<span class="o">=</span><span class="s1">'c -keys=cpu -link-params=0 -model=host'</span> <span class="se">\</span>
- --runtime<span class="o">=</span>crt <span class="se">\</span>
- --runtime-crt-system-lib <span class="m">1</span> <span class="se">\</span>
- --executor<span class="o">=</span><span class="s1">'graph'</span> <span class="se">\</span>
- --executor-graph-link-params <span class="m">0</span> <span class="se">\</span>
- --output model.tar <span class="se">\</span>
- --output-format mlf <span class="se">\</span>
- --pass-config tir.disable_vectorize<span class="o">=</span><span class="m">1</span> <span class="se">\</span>
- --disabled-pass<span class="o">=</span>AlterOpLayout
+ --target<span class="o">=</span><span class="s1">'c -keys=cpu -link-params=0 -model=host'</span> <span class="se">\</span>
+ --runtime<span class="o">=</span>crt <span class="se">\</span>
+ --runtime-crt-system-lib <span class="m">1</span> <span class="se">\</span>
+ --executor<span class="o">=</span><span class="s1">'graph'</span> <span class="se">\</span>
+ --executor-graph-link-params <span class="m">0</span> <span class="se">\</span>
+ --output model.tar <span class="se">\</span>
+ --output-format mlf <span class="se">\</span>
+ --pass-config tir.disable_vectorize<span class="o">=</span><span class="m">1</span> <span class="se">\</span>
+ --disabled-pass<span class="o">=</span>AlterOpLayout
</pre></div>
</div>
<p>This will generate a <code class="docutils literal notranslate"><span class="pre">model.tar</span></code> file which contains TVM compiler output files. To run this command for
@@ -421,10 +421,10 @@ the target is <code class="docutils literal notranslate"><span class="pre">--tar
for the project to <code class="docutils literal notranslate"><span class="pre">create</span></code> subcommand along with project options. Project options for each
platform (Zephyr/Arduino) are defined in their Project API server file. To generate Zephyr project, run:</p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>tvmc micro create <span class="se">\</span>
- project <span class="se">\</span>
- model.tar <span class="se">\</span>
- zephyr <span class="se">\</span>
- --project-option <span class="nv">project_type</span><span class="o">=</span>host_driven <span class="nv">zephyr_board</span><span class="o">=</span>qemu_x86
+ project <span class="se">\</span>
+ model.tar <span class="se">\</span>
+ zephyr <span class="se">\</span>
+ --project-option <span class="nv">project_type</span><span class="o">=</span>host_driven <span class="nv">zephyr_board</span><span class="o">=</span>qemu_x86
</pre></div>
</div>
<p>This will generate a <code class="docutils literal notranslate"><span class="pre">Host-Driven</span></code> Zephyr project for <code class="docutils literal notranslate"><span class="pre">qemu_x86</span></code> Zephyr board. In Host-Driven template project,
@@ -442,9 +442,9 @@ device using an RPC mechanism. Read more about <a class="reference external" hre
<div><p>Next step is to build the Zephyr project which includes TVM generated code for running the tiny model, Zephyr
template code to run a model in Host-Driven mode and TVM runtime source/header files. To build the project:</p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>tvmc micro build <span class="se">\</span>
- project <span class="se">\</span>
- zephyr <span class="se">\</span>
- --project-option <span class="nv">zephyr_board</span><span class="o">=</span>qemu_x86
+ project <span class="se">\</span>
+ zephyr <span class="se">\</span>
+ --project-option <span class="nv">zephyr_board</span><span class="o">=</span>qemu_x86
</pre></div>
</div>
<p>This will build the project in <code class="docutils literal notranslate"><span class="pre">project</span></code> directory and generates binary files under <code class="docutils literal notranslate"><span class="pre">project/build</span></code>. To build
@@ -452,9 +452,9 @@ Zephyr project for a different Zephyr board, change <code class="docutils litera
<p>Next, we flash the Zephyr binary file to Zephyr device. For <code class="docutils literal notranslate"><span class="pre">qemu_x86</span></code> Zephyr board this step does not
actually perform any action since QEMU will be used, however you need this step for physical hardware.</p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>tvmc micro flash <span class="se">\</span>
- project <span class="se">\</span>
- zephyr <span class="se">\</span>
- --project-option <span class="nv">zephyr_board</span><span class="o">=</span>qemu_x86
+ project <span class="se">\</span>
+ zephyr <span class="se">\</span>
+ --project-option <span class="nv">zephyr_board</span><span class="o">=</span>qemu_x86
</pre></div>
</div>
</div></blockquote>
@@ -467,12 +467,13 @@ The Zephyr board is waiting for host to open a communication channel. MicroTVM d
using a serial communication (UART). To run the flashed model on the device using TVMC, we use <code class="docutils literal notranslate"><span class="pre">tvmc</span> <span class="pre">run</span></code> subcommand
and pass <code class="docutils literal notranslate"><span class="pre">--device</span> <span class="pre">micro</span></code> to specify the device type. This command will open a communication channel, set input
values using <code class="docutils literal notranslate"><span class="pre">Graph</span> <span class="pre">Executor</span></code> on host and run full model on the device. Then it gets output from the device.</p>
-<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>tvmc run <span class="se">\</span>
- --device micro <span class="se">\</span>
- project <span class="se">\</span>
- --project-option <span class="nv">zephyr_board</span><span class="o">=</span>qemu_x86 <span class="se">\</span>
- --fill-mode ones
- --print-top <span class="m">4</span>
+<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span> tvmc run <span class="se">\</span>
+ --device micro <span class="se">\</span>
+ project <span class="se">\</span>
+ --project-option <span class="nv">zephyr_board</span><span class="o">=</span>qemu_x86 <span class="se">\</span>
+ --fill-mode ones <span class="se">\</span>
+ --print-top <span class="m">4</span>
+
<span class="c1"># Output:</span>
<span class="c1">#</span>
<span class="c1"># INFO:__main__:b'[100%] [QEMU] CPU: qemu32,+nx,+pae\n'</span>
diff --git a/docs/how_to/work_with_microtvm/sg_execution_times.html b/docs/how_to/work_with_microtvm/sg_execution_times.html
index 878fa4b..1053e49 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -300,13 +300,13 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:48.468</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>00:46.529</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:44.158</strong>: <a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></li>
-<li><p><strong>00:03.716</strong>: <a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></li>
-<li><p><strong>00:00.202</strong>: <a class="reference internal" href="micro_reference_vm.html#sphx-glr-how-to-work-with-microtvm-micro-reference-vm-py"><span class="std std-ref">microTVM Reference Virtual Machines</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_reference_vm.py</span></code>)</p></li>
-<li><p><strong>00:00.196</strong>: <a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></li>
-<li><p><strong>00:00.195</strong>: <a class="reference internal" href="micro_tvmc.html#sphx-glr-how-to-work-with-microtvm-micro-tvmc-py"><span class="std std-ref">Executing a Tiny Model with TVMC Micro</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tvmc.py</span></code>)</p></li>
+<li><p><strong>00:42.449</strong>: <a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></li>
+<li><p><strong>00:03.571</strong>: <a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></li>
+<li><p><strong>00:00.173</strong>: <a class="reference internal" href="micro_reference_vm.html#sphx-glr-how-to-work-with-microtvm-micro-reference-vm-py"><span class="std std-ref">microTVM Reference Virtual Machines</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_reference_vm.py</span></code>)</p></li>
+<li><p><strong>00:00.168</strong>: <a class="reference internal" href="micro_tvmc.html#sphx-glr-how-to-work-with-microtvm-micro-tvmc-py"><span class="std std-ref">Executing a Tiny Model with TVMC Micro</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tvmc.py</span></code>)</p></li>
+<li><p><strong>00:00.168</strong>: <a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index 2502273..30da67a 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -300,11 +300,11 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-relay-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:04.646</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:04.552</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:02.572</strong>: <a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></li>
-<li><p><strong>00:01.855</strong>: <a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></li>
-<li><p><strong>00:00.219</strong>: <a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></li>
+<li><p><strong>00:02.495</strong>: <a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></li>
+<li><p><strong>00:01.874</strong>: <a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></li>
+<li><p><strong>00:00.183</strong>: <a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/work_with_schedules/schedule_primitives.html b/docs/how_to/work_with_schedules/schedule_primitives.html
index 7b2f5fa..d30132d 100644
--- a/docs/how_to/work_with_schedules/schedule_primitives.html
+++ b/docs/how_to/work_with_schedules/schedule_primitives.html
@@ -640,13 +640,13 @@ of computation of <cite>C</cite>.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {C: Buffer(C_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
- A: Buffer(A_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
+ buffers = {A: Buffer(A_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
+ C: Buffer(C_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
B: Buffer(B_2: Pointer(float32), float32, [(stride_2: int32*m)], [], type="auto")}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (i: int32, 0, m) {
- B[(i*stride_2)] = (A[(i*stride_1)] + 1f32)
- C[(i*stride)] = (B[(i*stride_2)]*2f32)
+ B[(i*stride_2)] = (A[(i*stride)] + 1f32)
+ C[(i*stride_1)] = (B[(i*stride_2)]*2f32)
}
}
</pre></div>
@@ -669,12 +669,12 @@ tensor is required.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A: Buffer(A_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
- C: Buffer(C_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
+ A: Buffer(A_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
B: Buffer(B_2: Pointer(float32), float32, [(stride_2: int32*m)], [], type="auto")}
buffer_map = {A_1: A, B_1: B, C_1: C} {
for (i: int32, 0, m) {
- C[(i*stride_1)] = ((A[(i*stride)] + 1f32)*2f32)
+ C[(i*stride)] = ((A[(i*stride_1)] + 1f32)*2f32)
}
}
</pre></div>
diff --git a/docs/how_to/work_with_schedules/sg_execution_times.html b/docs/how_to/work_with_schedules/sg_execution_times.html
index 4fc8a0e..9e7a8c7 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -300,16 +300,16 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:06.155</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:05.815</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:02.225</strong>: <a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></li>
-<li><p><strong>00:01.355</strong>: <a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></li>
-<li><p><strong>00:00.770</strong>: <a class="reference internal" href="scan.html#sphx-glr-how-to-work-with-schedules-scan-py"><span class="std std-ref">Scan and Recurrent Kernel</span></a> (<code class="docutils literal notranslate"><span class="pre">scan.py</span></code>)</p></li>
-<li><p><strong>00:00.766</strong>: <a class="reference internal" href="reduction.html#sphx-glr-how-to-work-with-schedules-reduction-py"><span class="std std-ref">Reduction</span></a> (<code class="docutils literal notranslate"><span class="pre">reduction.py</span></code>)</p></li>
-<li><p><strong>00:00.330</strong>: <a class="reference internal" href="extern_op.html#sphx-glr-how-to-work-with-schedules-extern-op-py"><span class="std std-ref">External Tensor Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">extern_op.py</span></code>)</p></li>
-<li><p><strong>00:00.244</strong>: <a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></li>
-<li><p><strong>00:00.240</strong>: <a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></li>
-<li><p><strong>00:00.225</strong>: <a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></li>
+<li><p><strong>00:02.190</strong>: <a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></li>
+<li><p><strong>00:01.277</strong>: <a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></li>
+<li><p><strong>00:00.731</strong>: <a class="reference internal" href="scan.html#sphx-glr-how-to-work-with-schedules-scan-py"><span class="std std-ref">Scan and Recurrent Kernel</span></a> (<code class="docutils literal notranslate"><span class="pre">scan.py</span></code>)</p></li>
+<li><p><strong>00:00.725</strong>: <a class="reference internal" href="reduction.html#sphx-glr-how-to-work-with-schedules-reduction-py"><span class="std std-ref">Reduction</span></a> (<code class="docutils literal notranslate"><span class="pre">reduction.py</span></code>)</p></li>
+<li><p><strong>00:00.287</strong>: <a class="reference internal" href="extern_op.html#sphx-glr-how-to-work-with-schedules-extern-op-py"><span class="std std-ref">External Tensor Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">extern_op.py</span></code>)</p></li>
+<li><p><strong>00:00.208</strong>: <a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></li>
+<li><p><strong>00:00.207</strong>: <a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></li>
+<li><p><strong>00:00.190</strong>: <a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index b090599..3c33cdf 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -544,12 +544,12 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A: Buffer(A_2: Pointer(float32), float32, [65536], []),
- C: Buffer(C_2: Pointer(float32), float32, [524288], []),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [524288], []),
+ A: Buffer(A_2: Pointer(float32), float32, [65536], []),
B: Buffer(B_2: Pointer(float32), float32, [32768], [])}
buffer_map = {A_1: A, B_1: B, C_1: C} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpmr8tne_a/input0.cc'
-source_filename = "/tmp/tmpmr8tne_a/input0.cc"
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpdnbja0_x/input0.cc'
+source_filename = "/tmp/tmpdnbja0_x/input0.cc"
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-pc-linux-gnu"
diff --git a/docs/how_to/work_with_schedules/tuple_inputs.html b/docs/how_to/work_with_schedules/tuple_inputs.html
index 9d512e3..5c7587b 100644
--- a/docs/how_to/work_with_schedules/tuple_inputs.html
+++ b/docs/how_to/work_with_schedules/tuple_inputs.html
@@ -425,17 +425,17 @@ with <code class="xref py py-func docutils literal notranslate"><span class="pre
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>@main = primfn(idx_1: handle, val_1: handle, T_2: handle, T_3: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {T_1: Buffer(T_4: Pointer(int32), int32, [(stride: int32*m: int32)], [], type="auto"),
- T: Buffer(T_5: Pointer(int32), int32, [(stride_1: int32*m)], [], type="auto"),
- idx: Buffer(idx_2: Pointer(int32), int32, [(stride_2: int32*m)], [], type="auto"),
+ buffers = {T: Buffer(T_4: Pointer(int32), int32, [(stride: int32*m: int32)], [], type="auto"),
+ idx: Buffer(idx_2: Pointer(int32), int32, [(stride_1: int32*m)], [], type="auto"),
+ T_1: Buffer(T_5: Pointer(int32), int32, [(stride_2: int32*m)], [], type="auto"),
val: Buffer(val_2: Pointer(int32), int32, [(stride_3: int32*m)], [], type="auto")}
buffer_map = {idx_1: idx, val_1: val, T_2: T, T_3: T_1} {
for (i: int32, 0, m) {
- T[(i*stride_1)] = -1
- T_1[(i*stride)] = -2147483648
+ T[(i*stride)] = -1
+ T_1[(i*stride_2)] = -2147483648
for (k: int32, 0, n: int32) {
- T[(i*stride_1)] = @tir.if_then_else((val[((i*stride_3) + (k*stride_4: int32))] <= T_1[(i*stride)]), T[(i*stride_1)], idx[((i*stride_2) + (k*stride_5: int32))], dtype=int32)
- T_1[(i*stride)] = @tir.if_then_else((val[((i*stride_3) + (k*stride_4))] <= T_1[(i*stride)]), T_1[(i*stride)], val[((i*stride_3) + (k*stride_4))], dtype=int32)
+ T[(i*stride)] = @tir.if_then_else((val[((i*stride_3) + (k*stride_4: int32))] <= T_1[(i*stride_2)]), T[(i*stride)], idx[((i*stride_1) + (k*stride_5: int32))], dtype=int32)
+ T_1[(i*stride_2)] = @tir.if_then_else((val[((i*stride_3) + (k*stride_4))] <= T_1[(i*stride_2)]), T_1[(i*stride_2)], val[((i*stride_3) + (k*stride_4))], dtype=int32)
}
}
}
@@ -468,19 +468,19 @@ in terms of operation.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>@main = primfn(A0_1: handle, A1_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
- buffers = {A0: Buffer(A0_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
- C: Buffer(C_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
+ buffers = {C: Buffer(C_2: Pointer(float32), float32, [(stride: int32*m: int32)], [], type="auto"),
+ A0: Buffer(A0_2: Pointer(float32), float32, [(stride_1: int32*m)], [], type="auto"),
A1: Buffer(A1_2: Pointer(float32), float32, [(stride_2: int32*m)], [], type="auto")}
buffer_map = {A0_1: A0, A1_1: A1, C_1: C} {
allocate(B.v0: Pointer(global float32), float32, [n: int32]), storage_scope = global;
allocate(B.v1: Pointer(global float32), float32, [n]), storage_scope = global;
for (i: int32, 0, m) {
for (j: int32, 0, n) {
- B.v0_1: Buffer(B.v0, float32, [n], [])[j] = (A0[((i*stride) + (j*stride_3: int32))] + 2f32)
- B.v1_1: Buffer(B.v1, float32, [n], [])[j] = (A0[((i*stride) + (j*stride_3))]*3f32)
+ B.v0_1: Buffer(B.v0, float32, [n], [])[j] = (A0[((i*stride_1) + (j*stride_3: int32))] + 2f32)
+ B.v1_1: Buffer(B.v1, float32, [n], [])[j] = (A0[((i*stride_1) + (j*stride_3))]*3f32)
}
for (j_1: int32, 0, n) {
- C[((i*stride_1) + (j_1*stride_4: int32))] = (A1[((i*stride_2) + (j_1*stride_5: int32))] + B.v0_1[j_1])
+ C[((i*stride) + (j_1*stride_4: int32))] = (A1[((i*stride_2) + (j_1*stride_5: int32))] + B.v0_1[j_1])
}
}
}
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index fa68366..670772b 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1713,7 +1713,7 @@ Can be the a function or the function name.</p></li>
<dl class="py function">
<dt class="sig sig-object py" id="tvm.auto_scheduler.auto_schedule">
-<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
<dd><p>THIS API IS DEPRECATED.</p>
<p>Run auto scheduling search for a task.</p>
<dl class="field-list simple">
@@ -1750,7 +1750,7 @@ the initial naive schedule (state).</p>
<dl class="py class">
<dt class="sig sig-object py" id="tvm.auto_scheduler.SketchPolicy">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
<dd><p>The search policy that searches in a hierarchical search space defined by sketches.
The policy randomly samples programs from the space defined by sketches and use evolutionary
search to fine-tune them.</p>
diff --git a/docs/reference/api/python/target.html b/docs/reference/api/python/target.html
index 0448079..3502eed 100644
--- a/docs/reference/api/python/target.html
+++ b/docs/reference/api/python/target.html
@@ -791,7 +791,10 @@ Otherwise, separate versions are used for codegen and sim. Not
all allowed cpu strings will be valid, simulator will throw an
error if invalid. Does not affect codegen.</p></li>
<li><p><strong>llvm_options</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.10)"><em>str</em></a><em> or </em><em>list of str</em><em> (</em><em>default: None</em><em>)</em>) – User defined compiler arguments.</p></li>
+<li><p><strong>use_qfloat</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><em>bool</em></a><em> (</em><em>default: True for cpu_ver >= v68</em><em>, </em><em>False otherwise</em><em>)</em>) – Whether to use QFloat HVX instructions.</p></li>
+<li><p><strong>use_ieee_fp</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><em>bool</em></a><em> (</em><em>default: False</em><em>)</em>) – Whether to use IEEE HVX instructions</p></li>
<li><p><strong>link_params</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><em>bool</em></a><em> (</em><em>default: False</em><em>)</em>) – Whether to link graph parameters into the LLVM module.</p></li>
+<li><p><strong>Note</strong> (<em>Floating point support in HVX requires LLVM 14+.</em>) – </p></li>
</ul>
</dd>
</dl>
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index 8b3d838..e9f7e03 100644
--- a/docs/reference/api/typedoc/classes/bytestreamreader.html
+++ b/docs/reference/api/typedoc/classes/bytestreamreader.html
@@ -119,7 +119,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/15c1f75b8/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/6eaa5e6f2/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -141,7 +141,7 @@
<div class="tsd-signature tsd-kind-icon">bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Uint8Array</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/15c1f75b8/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/6eaa5e6f2/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
</ul>
</aside>
</section>
@@ -151,7 +151,7 @@
<div class="tsd-signature tsd-kind-icon">offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/15c1f75b8/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/6eaa5e6f2/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
</ul>
</aside>
</section>
@@ -168,7 +168,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/15c1f75b8/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/6eaa5e6f2/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">Uint8Array</span></h4>
@@ -185,7 +185,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/15c1f75b8/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/6eaa5e6f2/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -202,7 +202,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
... 2730 lines suppressed ...