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 2022/07/26 10:50:20 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@ea6ea42757f275573391eb3ff67034b2749948ae)
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/tvm-site.git
The following commit(s) were added to refs/heads/asf-site by this push:
new 6e8860f17 deploying docs (apache/tvm@ea6ea42757f275573391eb3ff67034b2749948ae)
6e8860f17 is described below
commit 6e8860f17a9f7f313ef218c6f901573b23232fc3
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Tue Jul 26 10:50:14 2022 +0000
deploying docs (apache/tvm@ea6ea42757f275573391eb3ff67034b2749948ae)
---
.../how_to/compile_models/from_darknet.rst.txt | 2 +-
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_oneflow.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 | 16 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 2 +-
.../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 | 16 +-
.../optimize_operators/sg_execution_times.rst.txt | 8 +-
.../sg_execution_times.rst.txt | 14 +-
.../tune_conv2d_layer_cuda.rst.txt | 2899 +++++++++++++++-----
.../tune_network_cuda.rst.txt | 2 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 492 +---
.../tune_with_autotvm/sg_execution_times.rst.txt | 6 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 26 +-
.../work_with_microtvm/micro_autotune.rst.txt | 16 +-
.../how_to/work_with_microtvm/micro_train.rst.txt | 16 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 8 +-
.../work_with_relay/sg_execution_times.rst.txt | 8 +-
.../how_to/work_with_schedules/intrin_math.rst.txt | 2 +-
.../work_with_schedules/sg_execution_times.rst.txt | 18 +-
.../how_to/work_with_schedules/tensorize.rst.txt | 2 +-
.../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 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 14 +-
docs/_sources/tutorial/autotvm_matmul_x86.rst.txt | 20 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 54 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 20 +-
.../tutorial/tensor_expr_get_started.rst.txt | 49 +-
docs/commit_hash | 2 +-
docs/how_to/compile_models/from_darknet.html | 2 +-
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_oneflow.html | 15 +-
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 | 18 +-
docs/how_to/deploy_models/deploy_prequantized.html | 9 +-
.../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 | 37 +-
docs/how_to/deploy_models/sg_execution_times.html | 16 +-
.../extend_tvm/bring_your_own_datatypes.html | 2 +-
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 | 16 +-
.../optimize_operators/sg_execution_times.html | 8 +-
.../sg_execution_times.html | 14 +-
.../tune_conv2d_layer_cuda.html | 2899 +++++++++++++++-----
.../tune_with_autoscheduler/tune_network_cuda.html | 2 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 492 +---
.../tune_with_autotvm/sg_execution_times.html | 6 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 26 +-
docs/how_to/work_with_microtvm/micro_autotune.html | 16 +-
docs/how_to/work_with_microtvm/micro_train.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 8 +-
.../how_to/work_with_relay/sg_execution_times.html | 8 +-
docs/how_to/work_with_schedules/intrin_math.html | 2 +-
.../work_with_schedules/sg_execution_times.html | 18 +-
docs/how_to/work_with_schedules/tensorize.html | 2 +-
docs/reference/api/python/auto_scheduler.html | 4 +-
.../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/sg_execution_times.html | 6 +-
docs/topic/vta/tutorials/sg_execution_times.html | 6 +-
docs/tutorial/auto_scheduler_matmul_x86.html | 6 +-
docs/tutorial/autotvm_matmul_x86.html | 20 +-
docs/tutorial/autotvm_relay_x86.html | 258 +-
docs/tutorial/cross_compilation_and_rpc.html | 2 +-
docs/tutorial/intro_topi.html | 2 +-
docs/tutorial/sg_execution_times.html | 20 +-
docs/tutorial/tensor_expr_get_started.html | 45 +-
121 files changed, 5382 insertions(+), 2998 deletions(-)
diff --git a/docs/_sources/how_to/compile_models/from_darknet.rst.txt b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
index 4e155b9dd..5ec07e4ad 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -317,7 +317,7 @@ The process is no different from other examples.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 1.933 seconds)
+ **Total running time of the script:** ( 1 minutes 3.724 seconds)
.. _sphx_glr_download_how_to_compile_models_from_darknet.py:
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 7f8f785d2..276517013 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -115,7 +115,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip52e1b03f-7187-44e0-8eaf-9940e8260f8f from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip8f3694ae-a876-4feb-a97d-0081dca49141 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_oneflow.rst.txt b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
index 61db3da98..f965022e6 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -113,7 +113,7 @@ Load a pretrained OneFlow model and save model
.. code-block:: none
Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
-
0%| | 0.00/41.5M [00:00<?, ?B/s]
19%|#9 | 7.99M/41.5M [00:00<00:00, 50.9MB/s]
35%|###4 | 14.3M/41.5M [00:00<00:00, 56.3MB/s]
48%|####7 | 19.8M/41.5M [00:00<00:00, 45.9MB/s]
59%|#####8 | 24.4M/41.5M [00:00<00:00, 24.8MB/s]
77%|#######7 | 32.0M/41.5M [00:01<00:00, 28.7MB/s]
85%|########5 | 35.3M/41.5M [00:01<00:00, 27.2MB/s]
92%|#########2| 38.3M/41.5M [00:01<00:00, 26.9MB/s]
100%|##########| 41.5M/41.5M [00:01<00:00, 31.1MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
19%|#9 | 7.99M/41.5M [00:00<00:00, 53.4MB/s]
39%|###8 | 16.0M/41.5M [00:00<00:00, 43.5MB/s]
58%|#####7 | 24.0M/41.5M [00:00<00:00, 45.0MB/s]
68%|######8 | 28.4M/41.5M [00:00<00:00, 44.1MB/s]
78%|#######8 | 32.6M/41.5M [00:00<00:00, 43.4MB/s]
92%|#########2| 38.3M/41.5M [00:00<00:00, 44.5MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 44.7MB/s]
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 92372e287..ced3d127e 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -94,7 +94,7 @@ 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]
13%|#2 | 5.80M/44.7M [00:00<00:00, 60.8MB/s]
26%|##5 | 11.6M/44.7M [00:00<00:00, 58.5MB/s]
77%|#######6 | 34.2M/44.7M [00:00<00:00, 138MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 135MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
37%|###6 | 16.4M/44.7M [00:00<00:00, 172MB/s]
86%|########5 | 38.3M/44.7M [00:00<00:00, 206MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 207MB/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 2a2b703b6..5a30163df 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -423,7 +423,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 4.623 seconds)
+ **Total running time of the script:** ( 1 minutes 6.971 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 c1c3f9ba3..7fdff2507 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
@@ -5,26 +5,26 @@
Computation times
=================
-**05:04.184** total execution time for **how_to_compile_models** files:
+**05:11.910** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:04.623 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:06.971 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:01.933 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:03.724 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:39.601 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:41.165 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:28.488 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:27.598 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:25.761 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:25.967 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:25.337 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:25.928 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:21.884 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:22.799 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:19.605 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:20.387 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:14.552 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:14.788 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.401 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.583 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
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 ebc9b22d0..bf0d90be5 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
@@ -441,7 +441,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.5391 15.5075 15.7231 15.4456 0.0831
+ 16.1154 15.9764 16.7093 15.8759 0.3033
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 dd9387fdf..6620144e1 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
@@ -123,7 +123,7 @@ 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]
10%|# | 17.7M/170M [00:00<00:00, 186MB/s]
24%|##4 | 41.0M/170M [00:00<00:00, 220MB/s]
38%|###8 | 64.8M/170M [00:00<00:00, 233MB/s]
52%|#####2 | 89.1M/170M [00:00<00:00, 242MB/s]
67%|######7 | 114M/170M [00:00<00:00, 250MB/s]
81%|########1 | 138M/170M [00:00<00:00, 245MB/s]
97%|#########6| 165M/170M [00:00<00:00, 255MB/s]
100%|##########| 170M/170M [00:00<00:00, 245MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
9%|9 | 15.4M/170M [00:00<00:01, 161MB/s]
21%|##1 | 36.5M/170M [00:00<00:00, 196MB/s]
33%|###3 | 56.8M/170M [00:00<00:00, 204MB/s]
47%|####6 | 79.2M/170M [00:00<00:00, 216MB/s]
61%|###### | 103M/170M [00:00<00:00, 228MB/s]
75%|#######4 | 127M/170M [00:00<00:00, 235MB/s]
88%|########7 | 149M/170M [00:00<00:00, 236MB/s]
100%|##########| 170M/170M [00:00<00:00, 226MB/s]
/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:3878: 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).
for i in range(dim)
/usr/local/lib/python3.7/dist-packages/torchvision/models/detection/anchor_utils.py:127: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -292,7 +292,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 51.868 seconds)
+ **Total running time of the script:** ( 3 minutes 0.852 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 de236a5fc..720755c3b 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -232,7 +232,7 @@ 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]
30%|##9 | 4.01M/13.6M [00:00<00:00, 41.8MB/s]
59%|#####9 | 8.00M/13.6M [00:00<00:00, 32.1MB/s]
90%|######### | 12.2M/13.6M [00:00<00:00, 36.9MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 37.9MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 158MB/s]
@@ -412,7 +412,7 @@ 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)
- 90.2425 90.0816 95.8397 89.8729 0.6945
+ 90.4531 90.2741 95.8757 90.1546 0.8351
@@ -461,7 +461,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 6.952 seconds)
+ **Total running time of the script:** ( 1 minutes 10.607 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 71cee696e..60aefff5b 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
@@ -439,7 +439,7 @@ 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)
- 118.4942 118.5807 119.4968 116.9254 0.4745
+ 120.9951 120.9853 121.9960 120.0151 0.4267
@@ -476,7 +476,7 @@ 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 57.893 seconds)
+ **Total running time of the script:** ( 1 minutes 58.945 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 dfe2da168..6f46b5522 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -255,7 +255,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 35.038 seconds)
+ **Total running time of the script:** ( 1 minutes 36.686 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 2631f33a7..be2c942e3 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
@@ -158,7 +158,7 @@ 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%|4 | 6416/132723 [00:00<00:01, 64148.44KB/s]
11%|#1 | 15072/132723 [00:00<00:01, 77327.85KB/s]
17%|#7 | 22805/132723 [00:00<00:01, 67826.06KB/s]
24%|##3 | 31402/132723 [00:00<00:01, 74426.31KB/s]
29%|##9 | 38974/132723 [00:00<00:01, 55477.01KB/s]
36%|###5 | 47595/132723 [00:00<00:01, 63644.67KB/s]
42%|####2 | 56247/132723 [00:00<00:01, 69939.60KB/s]
49%|####8 | 64980/132723 [00:00<00:00, 74842.81KB/s]
55%|#####4 | 72908/132723 [00:01<00:01, 58960.98KB/s]
61%|######1 | 81556/132723 [00:01<00:00, 65598.66KB/s]
67%|######6 | 88844/132723 [00:01<00:00, 62712.22KB/s]
74%|#######3 | 97556/132723 [00:01<00:00, 68909.03KB/s]
79%|#######9 | 104923/132723 [00:01<00:00, 57903.20KB/s]
86%|########5 | 113589/132723 [00:01<00:00, 64768.32KB/s]
91%|######### | 120658/132723 [00:01<00:00, 55674.31KB/s]
97%|########
#7| 129350/132723 [00:02<00:00, 62981.16KB/s]
100%|##########| 132723/132723 [00:02<00:00, 64465.36KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
4%|4 | 5714/132723 [00:00<00:02, 57136.38KB/s]
10%|9 | 13185/132723 [00:00<00:01, 67470.45KB/s]
16%|#5 | 20825/132723 [00:00<00:01, 71543.85KB/s]
21%|##1 | 28474/132723 [00:00<00:01, 73495.14KB/s]
27%|##7 | 36085/132723 [00:00<00:01, 74434.34KB/s]
33%|###2 | 43675/132723 [00:00<00:01, 74930.05KB/s]
39%|###8 | 51196/132723 [00:00<00:01, 75020.07KB/s]
44%|####4 | 58809/132723 [00:00<00:00, 75371.70KB/s]
50%|##### | 66475/132723 [00:00<00:00, 75773.26KB/s]
56%|#####5 | 74161/132723 [00:01<00:00, 76106.02KB/s]
62%|######1 | 81873/132723 [00:01<00:00, 76412.95KB/s]
67%|######7 | 89515/132723 [00:01<00:00, 70234.51KB/s]
73%|#######3 | 97065/132723 [00:01<00:00, 71734.88KB/s]
79%|#######8 | 104609/132723 [00:01<00:00, 72806.05KB/s]
85%|########4 | 112289/132723 [00:01<00:00, 73971.53KB/s]
90%|#########
| 119731/132723 [00:01<00:00, 65394.26KB/s]
96%|#########6| 127456/132723 [00:01<00:00, 68605.27KB/s]
100%|##########| 132723/132723 [00:01<00:00, 71998.12KB/s]
@@ -241,7 +241,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 29.778 seconds)
+ **Total running time of the script:** ( 2 minutes 38.979 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 23b5bf25c..c49270b24 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
@@ -5,22 +5,22 @@
Computation times
=================
-**10:52.592** total execution time for **how_to_deploy_models** files:
+**11:19.030** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 02:51.868 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:00.852 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 02:29.778 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 02:38.979 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 01:57.893 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 01:58.945 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:35.038 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:36.686 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:06.952 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:10.607 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:28.604 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:29.824 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:22.454 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:23.130 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``) | 00:00.006 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
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 48c58b584..a30e4dd1c 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
@@ -476,7 +476,7 @@ 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.zip8a4ec15b-dd30-4708-9e57-4287dc08f05d from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip16beb6ff-04ac-4110-a3ad-46f0a7b37d55 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
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 54783d323..0c557b407 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
@@ -5,14 +5,14 @@
Computation times
=================
-**00:39.491** total execution time for **how_to_extend_tvm** files:
+**00:41.583** total execution time for **how_to_extend_tvm** files:
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:36.443 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:38.303 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.141 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.309 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:00.900 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:00.961 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.007 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.009 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
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 9c8bb9fca..a5a7ee687 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
@@ -216,10 +216,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 6404us [6404us] (45.59%; 45.59%)
- FoldScaleAxis: 7644us [5us] (54.41%; 54.41%)
- FoldConstant: 7639us [1587us] (54.38%; 99.94%)
- InferType: 6053us [6053us] (43.08%; 79.23%)
+ InferType: 6829us [6829us] (45.29%; 45.29%)
+ FoldScaleAxis: 8250us [7us] (54.71%; 54.71%)
+ FoldConstant: 8243us [1615us] (54.67%; 99.92%)
+ InferType: 6628us [6628us] (43.96%; 80.41%)
@@ -258,10 +258,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 6125us [6125us] (44.64%; 44.64%)
- FoldScaleAxis: 7596us [4us] (55.36%; 55.36%)
- FoldConstant: 7592us [1567us] (55.33%; 99.94%)
- InferType: 6025us [6025us] (43.91%; 79.36%)
+ InferType: 6643us [6643us] (44.83%; 44.83%)
+ FoldScaleAxis: 8176us [7us] (55.17%; 55.17%)
+ FoldConstant: 8169us [1704us] (55.13%; 99.92%)
+ InferType: 6466us [6466us] (43.63%; 79.15%)
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 26ff698f3..0cb740fc2 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
@@ -340,7 +340,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 38.155233 ms
+ Convolution: 54.298894 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 9c0fef4e5..01bb5a1e6 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
@@ -671,7 +671,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 7.812188 ms
+ conv2d with tensor core: 7.178832 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 d4c068be1..d12c27bd6 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -143,8 +143,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.018373
- Baseline: 3.191560
+ Numpy running time: 0.020718
+ Baseline: 3.453568
@@ -239,7 +239,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.289777
+ Opt1: 0.319622
@@ -342,7 +342,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.331577
+ Opt2: 0.345024
@@ -438,7 +438,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.117284
+ Opt3: 0.122990
@@ -563,7 +563,7 @@ flattening.
.. code-block:: none
- Opt4: 0.109667
+ Opt4: 0.111755
@@ -685,7 +685,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.110944
+ Opt5: 0.111711
@@ -810,7 +810,7 @@ Futhermore, we can also utilize multi-core processors to do the thread-level par
.. code-block:: none
- Opt6: 0.144450
+ Opt6: 0.145288
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 85abbc281..9b05ac2bc 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
@@ -5,12 +5,12 @@
Computation times
=================
-**00:33.734** total execution time for **how_to_optimize_operators** files:
+**00:35.361** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:31.404 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:33.031 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.294 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.279 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.036 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.050 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
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 dbe7bf07c..3c1062fb2 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
@@ -5,18 +5,18 @@
Computation times
=================
-**05:58.163** total execution time for **how_to_tune_with_autoscheduler** files:
+**06:09.006** total execution time for **how_to_tune_with_autoscheduler** files:
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 03:13.851 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 03:21.685 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:21.130 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:23.406 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 00:45.044 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 00:46.366 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:21.043 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:19.544 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:08.608 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:09.145 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:08.487 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:08.860 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
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 146c82c2a..20a552b9e 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
@@ -240,370 +240,1119 @@ cooperative fetching, unrolling and operator fusion.
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 64;
- allocate(conv2d_nchw: Pointer(local float32), float32, [4]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [504]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [192]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [4], [], 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, [28]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [324]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [1152]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [196], [], scope="local", align=32)[0] = 0f32
+ conv2d_nchw_1[14] = 0f32
conv2d_nchw_1[1] = 0f32
+ conv2d_nchw_1[15] = 0f32
conv2d_nchw_1[2] = 0f32
+ conv2d_nchw_1[16] = 0f32
conv2d_nchw_1[3] = 0f32
- for (rc.outer.outer: int32, 0, 64) {
- let cse_var_1: int32 = (rc.outer.outer*392)
+ conv2d_nchw_1[17] = 0f32
+ conv2d_nchw_1[4] = 0f32
+ conv2d_nchw_1[18] = 0f32
+ conv2d_nchw_1[5] = 0f32
+ conv2d_nchw_1[19] = 0f32
+ conv2d_nchw_1[6] = 0f32
+ conv2d_nchw_1[20] = 0f32
+ conv2d_nchw_1[7] = 0f32
+ conv2d_nchw_1[21] = 0f32
+ conv2d_nchw_1[8] = 0f32
+ conv2d_nchw_1[22] = 0f32
+ conv2d_nchw_1[9] = 0f32
+ conv2d_nchw_1[23] = 0f32
+ conv2d_nchw_1[10] = 0f32
+ conv2d_nchw_1[24] = 0f32
+ conv2d_nchw_1[11] = 0f32
+ conv2d_nchw_1[25] = 0f32
+ conv2d_nchw_1[12] = 0f32
+ conv2d_nchw_1[26] = 0f32
+ conv2d_nchw_1[13] = 0f32
+ conv2d_nchw_1[27] = 0f32
+ for (rc.outer.outer: int32, 0, 128) {
+ let cse_var_2: int32 = (rc.outer.outer*196)
+ let cse_var_1: int32 = (rc.outer.outer*36)
{
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [504], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (1 <= floormod(threadIdx.x_1, 7))), data[(((cse_var_1 + (floordiv(threadIdx.x_1, 63)*49)) + floormod(threadIdx.x_1, 63)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 1), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 294), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_1 < 14), dtype=bool) {
- pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else(((threadIdx.x_1 < 7) && (1 <= floormod(threadIdx.x_1, 7))), data[(((cse_var_1 + (floordiv((threadIdx.x_1 + 490), 63)*49)) + threadIdx.x_1) + 41)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [324], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else((((9 <= threadIdx.x_1) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[(((cse_var_2 + (floordiv(threadIdx.x_1, 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 56)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 56), 81)) && (floormod((threadIdx.x_1 + 56), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 56), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 56), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 31), 81)) && (floormod((threadIdx.x_1 + 31), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 112), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 31), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 168)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 6), 81)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 168), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 6), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 62), 81)) && (floormod((threadIdx.x_1 + 62), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 224), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 62), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ if @tir.likely((threadIdx.x_1 < 44), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 280)] = @tir.if_then_else((((threadIdx.x_1 < 35) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 280), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 37), 81), 9)*7)) + floormod((threadIdx.x_1 + 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, [192], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 24)*3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_2 < 94), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 98), 24)*4608)) + (rc.outer.outer*72)) + (floormod((threadIdx.x_2 + 2), 24)*3))]
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1: Buffer(kernel.shared, float32, [1152], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 36)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 56), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 20), 36), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 112), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 4), 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 168), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 24), 36), 9)*9)) + floormod((threadIdx.x_2 + 6), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 224), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 280)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 280), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 28), 36), 9)*9)) + floormod((threadIdx.x_2 + 1), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 336), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 12), 36), 9)*9)) + floormod((threadIdx.x_2 + 3), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 392), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 36), 9)*9)) + floormod((threadIdx.x_2 + 5), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 448), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 36), 9)*9)) + floormod((threadIdx.x_2 + 7), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 504)] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 36)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 36)) + 64512)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 560), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 20), 36), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 616)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 616), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 4), 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 672), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 24), 36), 9)*9)) + floormod((threadIdx.x_2 + 6), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 728)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 728), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 784), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 28), 36), 9)*9)) + floormod((threadIdx.x_2 + 1), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 840)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 840), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 12), 36), 9)*9)) + floormod((threadIdx.x_2 + 3), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 896), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 36), 9)*9)) + floormod((threadIdx.x_2 + 5), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 952)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 952), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 36), 9)*9)) + floormod((threadIdx.x_2 + 7), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1008)] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 36)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 36)) + 129024)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1064)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1064), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 20), 36), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1120), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 4), 36))]
}
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 95)]))
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)), data[(((cse_var_1 + (floordiv(threadIdx.x_1, 63)*49)) + floormod(threadIdx.x_1, 63)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 1), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 294), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_1 < 14), dtype=bool) {
- pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else((threadIdx.x_1 < 7), data[(((cse_var_1 + (floordiv((threadIdx.x_1 + 490), 63)*49)) + threadIdx.x_1) + 42)], 0f32, dtype=float32)
- }
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 24)*3)) + 1)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_2 < 94), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 98), 24)*4608)) + (rc.outer.outer*72)) + (floormod((threadIdx.x_2 + 2), 24)*3)) + 1)]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 95)]))
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (floormod(threadIdx.x_1, 7) < 6)), data[(((cse_var_1 + (floordiv(threadIdx.x_1, 63)*49)) + floormod(threadIdx.x_1, 63)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 1), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 294), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_1 < 14), dtype=bool) {
- pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else(((threadIdx.x_1 < 7) && (floormod(threadIdx.x_1, 7) < 6)), data[(((cse_var_1 + (floordiv((threadIdx.x_1 + 490), 63)*49)) + threadIdx.x_1) + 43)], 0f32, dtype=float32)
- }
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 24)*3)) + 2)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_2 < 94), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 98), 24)*4608)) + (rc.outer.outer*72)) + (floormod((threadIdx.x_2 + 2), 24)*3)) + 2)]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 95)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
}
}
- for (i1.inner: int32, 0, 4) {
- compute[((((blockIdx.x*392) + (floordiv(threadIdx.x, 49)*196)) + (i1.inner*49)) + floormod(threadIdx.x, 49))] = max((conv2d_nchw_1[i1.inner] + bias[(((blockIdx.x*8) + (floordiv(threadIdx.x, 49)*4)) + i1.inner)]), 0f32)
+ for (i1.inner: int32, 0, 2) {
+ for (i2.inner: int32, 0, 7) {
+ let cse_var_3: int32 = ((i1.inner*7) + i2.inner)
+ {
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[cse_var_3] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner)]), 0f32)
+ compute[((((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7)) + 784)] = max((conv2d_nchw_1[(cse_var_3 + 14)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner) + 16)]), 0f32)
+ }
+ }
}
}
}
@@ -658,7 +1407,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.247 ms
+ Execution time of this operator: 0.377 ms
@@ -707,32 +1456,32 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
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=4)
- 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_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=8)
+ 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=2)
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=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_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_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=1)
+ conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=1)
+ 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_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=4)
- 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=1)
- compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
+ compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+ compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=8)
+ compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
+ 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_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)
@@ -755,12 +1504,12 @@ 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=98)
+ 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=56)
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=1)
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=98)
+ 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=56)
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)
@@ -780,348 +1529,1086 @@ 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__(98) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[4];
- __shared__ float pad_temp_shared[504];
- __shared__ float kernel_shared[192];
+ extern "C" __global__ void __launch_bounds__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[28];
+ __shared__ float pad_temp_shared[324];
+ __shared__ float kernel_shared[1152];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[16] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
+ conv2d_nchw[17] = 0.000000e+00f;
+ conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[18] = 0.000000e+00f;
+ conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[19] = 0.000000e+00f;
+ conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[20] = 0.000000e+00f;
+ conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[21] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[22] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[23] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[24] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[25] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[26] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
+ conv2d_nchw[27] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 128; ++rc_outer_outer) {
__syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = ((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && (1 <= (((int)threadIdx.x) % 7))) ? data[((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 63) * 49)) + (((int)threadIdx.x) % 63)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 98)] = ((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 196)] = ((((1 <= (((((int)threadIdx.x) / 7) + 1) % 9)) && ((((((int)threadIdx.x) / 7) + 1) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 1) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 294)] = ((((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 294) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = ((((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 14) {
- pad_temp_shared[(((int)threadIdx.x) + 490)] = (((((int)threadIdx.x) < 7) && (1 <= (((int)threadIdx.x) % 7))) ? data[((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 490) / 63) * 49)) + ((int)threadIdx.x)) + 41)] : 0.000000e+00f);
+ pad_temp_shared[((int)threadIdx.x)] = ((((9 <= ((int)threadIdx.x)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[((((rc_outer_outer * 196) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 56)] = (((((9 <= ((((int)threadIdx.x) + 56) % 81)) && (((((int)threadIdx.x) + 56) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 56) / 81) * 49)) + ((((((int)threadIdx.x) + 56) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((9 <= ((((int)threadIdx.x) + 31) % 81)) && (((((int)threadIdx.x) + 31) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 112) / 81) * 49)) + ((((((int)threadIdx.x) + 31) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 168)] = ((((3 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 168) / 81) * 49)) + (((((int)threadIdx.x) + 6) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((9 <= ((((int)threadIdx.x) + 62) % 81)) && (((((int)threadIdx.x) + 62) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 224) / 81) * 49)) + ((((((int)threadIdx.x) + 62) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 44) {
+ pad_temp_shared[(((int)threadIdx.x) + 280)] = ((((((int)threadIdx.x) < 35) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 280) / 81) * 49)) + (((((int)threadIdx.x) + 37) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
}
- kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3))];
- if (((int)threadIdx.x) < 94) {
- kernel_shared[(((int)threadIdx.x) + 98)] = kernel[((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 98) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 2) % 24) * 3))];
+ kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 36) * 4608)) + (rc_outer_outer * 36)) + (((int)threadIdx.x) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 56)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 20) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 2) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) + 4) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 24) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 6) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) + 8) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 280)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 28) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 1) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 12) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 3) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 32) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 5) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 16) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 7) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 504)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 36) * 4608)) + (rc_outer_outer * 36)) + (((int)threadIdx.x) % 36)) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 20) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 2) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 616)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 616) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) + 4) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 672) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 24) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 6) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 728)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 728) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) + 8) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 28) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 1) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 840)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 840) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 12) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 3) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 896) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 32) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 5) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 952) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 16) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 7) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 36) * 4608)) + (rc_outer_outer * 36)) + (((int)threadIdx.x) % 36)) + 129024)];
+ kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1064) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 20) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 2) % 9))];
+ if (((int)threadIdx.x) < 32) {
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1120) / 36) * 4608)) + (rc_outer_outer * 36)) + (((int)threadIdx.x) + 4))];
}
__syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 95)]));
- __syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = (((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) ? data[((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 63) * 49)) + (((int)threadIdx.x) % 63)) - 7)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 98)] = (((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 196)] = (((1 <= (((((int)threadIdx.x) / 7) + 1) % 9)) && ((((((int)threadIdx.x) / 7) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 1) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 294)] = (((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 294) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = (((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 14) {
- pad_temp_shared[(((int)threadIdx.x) + 490)] = ((((int)threadIdx.x) < 7) ? data[((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 490) / 63) * 49)) + ((int)threadIdx.x)) + 42)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + 1)];
- if (((int)threadIdx.x) < 94) {
- kernel_shared[(((int)threadIdx.x) + 98)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 98) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 2) % 24) * 3)) + 1)];
- }
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 95)]));
- __syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = ((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && ((((int)threadIdx.x) % 7) < 6)) ? data[((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 63) * 49)) + (((int)threadIdx.x) % 63)) - 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 98)] = ((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 196)] = ((((1 <= (((((int)threadIdx.x) / 7) + 1) % 9)) && ((((((int)threadIdx.x) / 7) + 1) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 1) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 294)] = ((((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 294) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = ((((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 14) {
- pad_temp_shared[(((int)threadIdx.x) + 490)] = (((((int)threadIdx.x) < 7) && ((((int)threadIdx.x) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 490) / 63) * 49)) + ((int)threadIdx.x)) + 43)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + 2)];
- if (((int)threadIdx.x) < 94) {
- kernel_shared[(((int)threadIdx.x) + 98)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 98) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 2) % 24) * 3)) + 2)];
- }
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 95)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
}
- for (int i1_inner = 0; i1_inner < 4; ++i1_inner) {
- compute[((((((int)blockIdx.x) * 392) + ((((int)threadIdx.x) / 49) * 196)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 8) + ((((int)threadIdx.x) / 49) * 4)) + i1_inner)]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+ for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ compute[((((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7)) + 784)] = max((conv2d_nchw[(((i1_inner * 7) + i2_inner) + 14)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner) + 16)]), 0.000000e+00f);
+ }
}
}
@@ -1183,7 +2670,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 13.851 seconds)
+ **Total running time of the script:** ( 3 minutes 21.685 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 fa9c803e4..fbbaeec46 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
@@ -647,7 +647,7 @@ 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)
- 9.8203 9.7996 9.8876 9.7736 0.0488
+ 9.6666 9.6901 9.7160 9.5937 0.0526
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 b4a7111c3..23c3ff467 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
@@ -666,7 +666,7 @@ 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)
- 751.6063 751.6842 751.8405 751.2941 0.2297
+ 763.3527 763.4943 764.3378 762.2261 0.8679
@@ -694,7 +694,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 21.130 seconds)
+ **Total running time of the script:** ( 1 minutes 23.406 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 fcef29529..1821113a9 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
@@ -397,407 +397,103 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_9: placeholder_15: Buffer(placeholder_14, float32, [128, 512], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_7: placeholder_16: Buffer(placeholder_12, int32, [4916], []), placeholder_8: placeholder_17: Buffer(placeholder_13, int32, [33], []), placeholder_5: placeholder_18: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 256) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [256]), storage_scope = global {
- for (nb_j.inner: int32, 0, 2) {
- let cse_var_2: int32 = (nb_j.inner*16)
- let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
- {
- compute_5: Buffer(compute_4, float32, [256], [])[cse_var_2] = 0f32
- compute_5[(cse_var_2 + 1)] = 0f32
- compute_5[(cse_var_2 + 2)] = 0f32
- compute_5[(cse_var_2 + 3)] = 0f32
- compute_5[(cse_var_2 + 4)] = 0f32
- compute_5[(cse_var_2 + 5)] = 0f32
- compute_5[(cse_var_2 + 6)] = 0f32
- compute_5[(cse_var_2 + 7)] = 0f32
- compute_5[(cse_var_2 + 8)] = 0f32
- compute_5[(cse_var_2 + 9)] = 0f32
- compute_5[(cse_var_2 + 10)] = 0f32
- compute_5[(cse_var_2 + 11)] = 0f32
- compute_5[(cse_var_2 + 12)] = 0f32
- compute_5[(cse_var_2 + 13)] = 0f32
- compute_5[(cse_var_2 + 14)] = 0f32
- compute_5[(cse_var_2 + 15)] = 0f32
- compute_5[(cse_var_2 + 32)] = 0f32
- compute_5[(cse_var_2 + 33)] = 0f32
- compute_5[(cse_var_2 + 34)] = 0f32
- compute_5[(cse_var_2 + 35)] = 0f32
- compute_5[(cse_var_2 + 36)] = 0f32
- compute_5[(cse_var_2 + 37)] = 0f32
- compute_5[(cse_var_2 + 38)] = 0f32
- compute_5[(cse_var_2 + 39)] = 0f32
- compute_5[(cse_var_2 + 40)] = 0f32
- compute_5[(cse_var_2 + 41)] = 0f32
- compute_5[(cse_var_2 + 42)] = 0f32
- compute_5[(cse_var_2 + 43)] = 0f32
- compute_5[(cse_var_2 + 44)] = 0f32
- compute_5[(cse_var_2 + 45)] = 0f32
- compute_5[(cse_var_2 + 46)] = 0f32
- compute_5[(cse_var_2 + 47)] = 0f32
- compute_5[(cse_var_2 + 64)] = 0f32
- compute_5[(cse_var_2 + 65)] = 0f32
- compute_5[(cse_var_2 + 66)] = 0f32
- compute_5[(cse_var_2 + 67)] = 0f32
- compute_5[(cse_var_2 + 68)] = 0f32
- compute_5[(cse_var_2 + 69)] = 0f32
- compute_5[(cse_var_2 + 70)] = 0f32
- compute_5[(cse_var_2 + 71)] = 0f32
- compute_5[(cse_var_2 + 72)] = 0f32
- compute_5[(cse_var_2 + 73)] = 0f32
- compute_5[(cse_var_2 + 74)] = 0f32
- compute_5[(cse_var_2 + 75)] = 0f32
- compute_5[(cse_var_2 + 76)] = 0f32
- compute_5[(cse_var_2 + 77)] = 0f32
- compute_5[(cse_var_2 + 78)] = 0f32
- compute_5[(cse_var_2 + 79)] = 0f32
- compute_5[(cse_var_2 + 96)] = 0f32
- compute_5[(cse_var_2 + 97)] = 0f32
- compute_5[(cse_var_2 + 98)] = 0f32
- compute_5[(cse_var_2 + 99)] = 0f32
- compute_5[(cse_var_2 + 100)] = 0f32
- compute_5[(cse_var_2 + 101)] = 0f32
- compute_5[(cse_var_2 + 102)] = 0f32
- compute_5[(cse_var_2 + 103)] = 0f32
- compute_5[(cse_var_2 + 104)] = 0f32
- compute_5[(cse_var_2 + 105)] = 0f32
- compute_5[(cse_var_2 + 106)] = 0f32
- compute_5[(cse_var_2 + 107)] = 0f32
- compute_5[(cse_var_2 + 108)] = 0f32
- compute_5[(cse_var_2 + 109)] = 0f32
- compute_5[(cse_var_2 + 110)] = 0f32
- compute_5[(cse_var_2 + 111)] = 0f32
- compute_5[(cse_var_2 + 128)] = 0f32
- compute_5[(cse_var_2 + 129)] = 0f32
- compute_5[(cse_var_2 + 130)] = 0f32
- compute_5[(cse_var_2 + 131)] = 0f32
- compute_5[(cse_var_2 + 132)] = 0f32
- compute_5[(cse_var_2 + 133)] = 0f32
- compute_5[(cse_var_2 + 134)] = 0f32
- compute_5[(cse_var_2 + 135)] = 0f32
- compute_5[(cse_var_2 + 136)] = 0f32
- compute_5[(cse_var_2 + 137)] = 0f32
- compute_5[(cse_var_2 + 138)] = 0f32
- compute_5[(cse_var_2 + 139)] = 0f32
- compute_5[(cse_var_2 + 140)] = 0f32
- compute_5[(cse_var_2 + 141)] = 0f32
- compute_5[(cse_var_2 + 142)] = 0f32
- compute_5[(cse_var_2 + 143)] = 0f32
- compute_5[(cse_var_2 + 160)] = 0f32
- compute_5[(cse_var_2 + 161)] = 0f32
- compute_5[(cse_var_2 + 162)] = 0f32
- compute_5[(cse_var_2 + 163)] = 0f32
- compute_5[(cse_var_2 + 164)] = 0f32
- compute_5[(cse_var_2 + 165)] = 0f32
- compute_5[(cse_var_2 + 166)] = 0f32
- compute_5[(cse_var_2 + 167)] = 0f32
- compute_5[(cse_var_2 + 168)] = 0f32
- compute_5[(cse_var_2 + 169)] = 0f32
- compute_5[(cse_var_2 + 170)] = 0f32
- compute_5[(cse_var_2 + 171)] = 0f32
- compute_5[(cse_var_2 + 172)] = 0f32
- compute_5[(cse_var_2 + 173)] = 0f32
- compute_5[(cse_var_2 + 174)] = 0f32
- compute_5[(cse_var_2 + 175)] = 0f32
- compute_5[(cse_var_2 + 192)] = 0f32
- compute_5[(cse_var_2 + 193)] = 0f32
- compute_5[(cse_var_2 + 194)] = 0f32
- compute_5[(cse_var_2 + 195)] = 0f32
- compute_5[(cse_var_2 + 196)] = 0f32
- compute_5[(cse_var_2 + 197)] = 0f32
- compute_5[(cse_var_2 + 198)] = 0f32
- compute_5[(cse_var_2 + 199)] = 0f32
- compute_5[(cse_var_2 + 200)] = 0f32
- compute_5[(cse_var_2 + 201)] = 0f32
- compute_5[(cse_var_2 + 202)] = 0f32
- compute_5[(cse_var_2 + 203)] = 0f32
- compute_5[(cse_var_2 + 204)] = 0f32
- compute_5[(cse_var_2 + 205)] = 0f32
- compute_5[(cse_var_2 + 206)] = 0f32
- compute_5[(cse_var_2 + 207)] = 0f32
- compute_5[(cse_var_2 + 224)] = 0f32
- compute_5[(cse_var_2 + 225)] = 0f32
- compute_5[(cse_var_2 + 226)] = 0f32
- compute_5[(cse_var_2 + 227)] = 0f32
- compute_5[(cse_var_2 + 228)] = 0f32
- compute_5[(cse_var_2 + 229)] = 0f32
- compute_5[(cse_var_2 + 230)] = 0f32
- compute_5[(cse_var_2 + 231)] = 0f32
- compute_5[(cse_var_2 + 232)] = 0f32
- compute_5[(cse_var_2 + 233)] = 0f32
- compute_5[(cse_var_2 + 234)] = 0f32
- compute_5[(cse_var_2 + 235)] = 0f32
- compute_5[(cse_var_2 + 236)] = 0f32
- compute_5[(cse_var_2 + 237)] = 0f32
- compute_5[(cse_var_2 + 238)] = 0f32
- compute_5[(cse_var_2 + 239)] = 0f32
- for (elem_idx: int32, 0, (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- let cse_var_131: int32 = (elem_idx*16)
- let cse_var_130: int32 = (cse_var_2 + 99)
- let cse_var_129: int32 = (cse_var_2 + 98)
- let cse_var_128: int32 = (cse_var_2 + 97)
- let cse_var_127: int32 = (cse_var_2 + 96)
- let cse_var_126: int32 = (cse_var_2 + 9)
- let cse_var_125: int32 = (cse_var_2 + 8)
- let cse_var_124: int32 = (cse_var_2 + 79)
- let cse_var_123: int32 = (cse_var_2 + 78)
- let cse_var_122: int32 = (cse_var_2 + 77)
- let cse_var_121: int32 = (cse_var_2 + 76)
- let cse_var_120: int32 = (cse_var_2 + 75)
- let cse_var_119: int32 = (cse_var_2 + 74)
- let cse_var_118: int32 = (cse_var_2 + 73)
- let cse_var_117: int32 = (cse_var_2 + 72)
- let cse_var_116: int32 = (cse_var_2 + 71)
- let cse_var_115: int32 = (cse_var_2 + 70)
- let cse_var_114: int32 = (cse_var_2 + 7)
- let cse_var_113: int32 = (cse_var_2 + 69)
- let cse_var_112: int32 = (cse_var_2 + 68)
- let cse_var_111: int32 = (cse_var_2 + 67)
- let cse_var_110: int32 = (cse_var_2 + 66)
- let cse_var_109: int32 = (cse_var_2 + 65)
- let cse_var_108: int32 = (cse_var_2 + 64)
- let cse_var_107: int32 = (cse_var_2 + 6)
- let cse_var_106: int32 = (cse_var_2 + 5)
- let cse_var_105: int32 = (cse_var_2 + 47)
- let cse_var_104: int32 = (cse_var_2 + 46)
- let cse_var_103: int32 = (cse_var_2 + 45)
- let cse_var_102: int32 = (cse_var_2 + 44)
- let cse_var_101: int32 = (cse_var_2 + 43)
- let cse_var_100: int32 = (cse_var_2 + 42)
- let cse_var_99: int32 = (cse_var_2 + 41)
- let cse_var_98: int32 = (cse_var_2 + 40)
- let cse_var_97: int32 = (cse_var_2 + 4)
- let cse_var_96: int32 = (cse_var_2 + 39)
- let cse_var_95: int32 = (cse_var_2 + 38)
- let cse_var_94: int32 = (cse_var_2 + 37)
- let cse_var_93: int32 = (cse_var_2 + 36)
- let cse_var_92: int32 = (cse_var_2 + 35)
- let cse_var_91: int32 = (cse_var_2 + 34)
- let cse_var_90: int32 = (cse_var_2 + 33)
- let cse_var_89: int32 = (cse_var_2 + 32)
- let cse_var_88: int32 = (cse_var_2 + 3)
- let cse_var_87: int32 = (cse_var_2 + 239)
- let cse_var_86: int32 = (cse_var_2 + 238)
- let cse_var_85: int32 = (cse_var_2 + 237)
- let cse_var_84: int32 = (cse_var_2 + 236)
- let cse_var_83: int32 = (cse_var_2 + 235)
- let cse_var_82: int32 = (cse_var_2 + 234)
- let cse_var_81: int32 = (cse_var_2 + 233)
- let cse_var_80: int32 = (cse_var_2 + 232)
- let cse_var_79: int32 = (cse_var_2 + 231)
- let cse_var_78: int32 = (cse_var_2 + 230)
- let cse_var_77: int32 = (cse_var_2 + 229)
- let cse_var_76: int32 = (cse_var_2 + 228)
- let cse_var_75: int32 = (cse_var_2 + 227)
- let cse_var_74: int32 = (cse_var_2 + 226)
- let cse_var_73: int32 = (cse_var_2 + 225)
- let cse_var_72: int32 = (cse_var_2 + 224)
- let cse_var_71: int32 = (cse_var_2 + 207)
- let cse_var_70: int32 = (cse_var_2 + 206)
- let cse_var_69: int32 = (cse_var_2 + 205)
- let cse_var_68: int32 = (cse_var_2 + 204)
- let cse_var_67: int32 = (cse_var_2 + 203)
- let cse_var_66: int32 = (cse_var_2 + 202)
- let cse_var_65: int32 = (cse_var_2 + 201)
- let cse_var_64: int32 = (cse_var_2 + 200)
- let cse_var_63: int32 = (cse_var_2 + 2)
- let cse_var_62: int32 = (cse_var_2 + 199)
- let cse_var_61: int32 = (cse_var_2 + 198)
- let cse_var_60: int32 = (cse_var_2 + 197)
- let cse_var_59: int32 = (cse_var_2 + 196)
- let cse_var_58: int32 = (cse_var_2 + 195)
- let cse_var_57: int32 = (cse_var_2 + 194)
- let cse_var_56: int32 = (cse_var_2 + 193)
- let cse_var_55: int32 = (cse_var_2 + 192)
- let cse_var_54: int32 = (cse_var_2 + 175)
- let cse_var_53: int32 = (cse_var_2 + 174)
- let cse_var_52: int32 = (cse_var_2 + 173)
- let cse_var_51: int32 = (cse_var_2 + 172)
- let cse_var_50: int32 = (cse_var_2 + 171)
- let cse_var_49: int32 = (cse_var_2 + 170)
- let cse_var_48: int32 = (cse_var_2 + 169)
- let cse_var_47: int32 = (cse_var_2 + 168)
- let cse_var_46: int32 = (cse_var_2 + 167)
- let cse_var_45: int32 = (cse_var_2 + 166)
- let cse_var_44: int32 = (cse_var_2 + 165)
- let cse_var_43: int32 = (cse_var_2 + 164)
- let cse_var_42: int32 = (cse_var_2 + 163)
- let cse_var_41: int32 = (cse_var_2 + 162)
- let cse_var_40: int32 = (cse_var_2 + 161)
- let cse_var_39: int32 = (cse_var_2 + 160)
- let cse_var_38: int32 = (cse_var_2 + 15)
- let cse_var_37: int32 = (cse_var_2 + 143)
- let cse_var_36: int32 = (cse_var_2 + 142)
- let cse_var_35: int32 = (cse_var_2 + 141)
- let cse_var_34: int32 = (cse_var_2 + 140)
- let cse_var_33: int32 = (cse_var_2 + 14)
- let cse_var_32: int32 = (cse_var_2 + 139)
- let cse_var_31: int32 = (cse_var_2 + 138)
- let cse_var_30: int32 = (cse_var_2 + 137)
- let cse_var_29: int32 = (cse_var_2 + 136)
- let cse_var_28: int32 = (cse_var_2 + 135)
- let cse_var_27: int32 = (cse_var_2 + 134)
- let cse_var_26: int32 = (cse_var_2 + 133)
- let cse_var_25: int32 = (cse_var_2 + 132)
- let cse_var_24: int32 = (cse_var_2 + 131)
- let cse_var_23: int32 = (cse_var_2 + 130)
- let cse_var_22: int32 = (cse_var_2 + 13)
- let cse_var_21: int32 = (cse_var_2 + 129)
- let cse_var_20: int32 = (cse_var_2 + 128)
- let cse_var_19: int32 = (cse_var_2 + 12)
- let cse_var_18: int32 = (cse_var_2 + 111)
- let cse_var_17: int32 = (cse_var_2 + 110)
- let cse_var_16: int32 = (cse_var_2 + 11)
- let cse_var_15: int32 = (cse_var_2 + 109)
- let cse_var_14: int32 = (cse_var_2 + 108)
- let cse_var_13: int32 = (cse_var_2 + 107)
- let cse_var_12: int32 = (cse_var_2 + 106)
- let cse_var_11: int32 = (cse_var_2 + 105)
- let cse_var_10: int32 = (cse_var_2 + 104)
- let cse_var_9: int32 = (cse_var_2 + 103)
- let cse_var_8: int32 = (cse_var_2 + 102)
- let cse_var_7: int32 = (cse_var_2 + 101)
- let cse_var_6: int32 = (cse_var_2 + 100)
- let cse_var_5: int32 = (cse_var_2 + 10)
- let cse_var_4: int32 = (cse_var_2 + 1)
- let cse_var_3: int32 = (floordiv(i0.outer.i1.outer.fused, 16)*2048)
- {
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_63] = (compute_5[cse_var_63] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_88] = (compute_5[cse_var_88] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_97] = (compute_5[cse_var_97] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_106] = (compute_5[cse_var_106] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_107] = (compute_5[cse_var_107] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_114] = (compute_5[cse_var_114] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_125] = (compute_5[cse_var_125] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_126] = (compute_5[cse_var_126] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_22] = (compute_5[cse_var_22] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_33] = (compute_5[cse_var_33] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_38] = (compute_5[cse_var_38] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_89] = (compute_5[cse_var_89] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_90] = (compute_5[cse_var_90] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_91] = (compute_5[cse_var_91] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_92] = (compute_5[cse_var_92] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_93] = (compute_5[cse_var_93] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_94] = (compute_5[cse_var_94] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_95] = (compute_5[cse_var_95] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_96] = (compute_5[cse_var_96] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_98] = (compute_5[cse_var_98] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_99] = (compute_5[cse_var_99] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_100] = (compute_5[cse_var_100] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_101] = (compute_5[cse_var_101] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_102] = (compute_5[cse_var_102] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_103] = (compute_5[cse_var_103] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_104] = (compute_5[cse_var_104] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_105] = (compute_5[cse_var_105] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_108] = (compute_5[cse_var_108] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_109] = (compute_5[cse_var_109] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_110] = (compute_5[cse_var_110] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_111] = (compute_5[cse_var_111] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_112] = (compute_5[cse_var_112] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_113] = (compute_5[cse_var_113] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_115] = (compute_5[cse_var_115] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_116] = (compute_5[cse_var_116] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_117] = (compute_5[cse_var_117] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_118] = (compute_5[cse_var_118] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_119] = (compute_5[cse_var_119] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_120] = (compute_5[cse_var_120] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_121] = (compute_5[cse_var_121] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_122] = (compute_5[cse_var_122] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_123] = (compute_5[cse_var_123] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_124] = (compute_5[cse_var_124] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_127] = (compute_5[cse_var_127] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_128] = (compute_5[cse_var_128] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_129] = (compute_5[cse_var_129] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_130] = (compute_5[cse_var_130] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_18] = (compute_5[cse_var_18] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_20] = (compute_5[cse_var_20] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_21] = (compute_5[cse_var_21] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_23] = (compute_5[cse_var_23] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_24] = (compute_5[cse_var_24] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_25] = (compute_5[cse_var_25] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_26] = (compute_5[cse_var_26] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_27] = (compute_5[cse_var_27] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_28] = (compute_5[cse_var_28] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_29] = (compute_5[cse_var_29] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_30] = (compute_5[cse_var_30] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_31] = (compute_5[cse_var_31] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_32] = (compute_5[cse_var_32] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_34] = (compute_5[cse_var_34] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_35] = (compute_5[cse_var_35] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_36] = (compute_5[cse_var_36] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_37] = (compute_5[cse_var_37] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_39] = (compute_5[cse_var_39] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_40] = (compute_5[cse_var_40] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_41] = (compute_5[cse_var_41] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_42] = (compute_5[cse_var_42] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_43] = (compute_5[cse_var_43] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_44] = (compute_5[cse_var_44] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_45] = (compute_5[cse_var_45] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_46] = (compute_5[cse_var_46] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_47] = (compute_5[cse_var_47] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_48] = (compute_5[cse_var_48] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_49] = (compute_5[cse_var_49] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_50] = (compute_5[cse_var_50] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_51] = (compute_5[cse_var_51] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_52] = (compute_5[cse_var_52] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_53] = (compute_5[cse_var_53] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_54] = (compute_5[cse_var_54] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_55] = (compute_5[cse_var_55] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_56] = (compute_5[cse_var_56] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_57] = (compute_5[cse_var_57] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_58] = (compute_5[cse_var_58] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_59] = (compute_5[cse_var_59] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_60] = (compute_5[cse_var_60] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_61] = (compute_5[cse_var_61] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_62] = (compute_5[cse_var_62] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_64] = (compute_5[cse_var_64] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_65] = (compute_5[cse_var_65] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_66] = (compute_5[cse_var_66] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_67] = (compute_5[cse_var_67] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_68] = (compute_5[cse_var_68] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_69] = (compute_5[cse_var_69] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_70] = (compute_5[cse_var_70] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_71] = (compute_5[cse_var_71] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_72] = (compute_5[cse_var_72] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_73] = (compute_5[cse_var_73] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_74] = (compute_5[cse_var_74] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_75] = (compute_5[cse_var_75] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_76] = (compute_5[cse_var_76] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_77] = (compute_5[cse_var_77] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_78] = (compute_5[cse_var_78] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_79] = (compute_5[cse_var_79] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_80] = (compute_5[cse_var_80] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_81] = (compute_5[cse_var_81] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_82] = (compute_5[cse_var_82] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_83] = (compute_5[cse_var_83] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_84] = (compute_5[cse_var_84] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_85] = (compute_5[cse_var_85] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_86] = (compute_5[cse_var_86] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_87] = (compute_5[cse_var_87] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ preflattened_buffer_map = {placeholder_6: placeholder_15: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_7: placeholder_16: Buffer(placeholder_12, int32, [4916], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_18: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_19: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 32) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [2048]), storage_scope = global {
+ for (i.outer.inner: int32, 0, 64) {
+ for (i.inner.init: int32, 0, 2) {
+ let cse_var_1: int32 = ((i.outer.inner*32) + (i.inner.init*16))
+ {
+ compute_5: Buffer(compute_4, float32, [2048], [])[cse_var_1] = 0f32
+ compute_5[(cse_var_1 + 1)] = 0f32
+ compute_5[(cse_var_1 + 2)] = 0f32
+ compute_5[(cse_var_1 + 3)] = 0f32
+ compute_5[(cse_var_1 + 4)] = 0f32
+ compute_5[(cse_var_1 + 5)] = 0f32
+ compute_5[(cse_var_1 + 6)] = 0f32
+ compute_5[(cse_var_1 + 7)] = 0f32
+ compute_5[(cse_var_1 + 8)] = 0f32
+ compute_5[(cse_var_1 + 9)] = 0f32
+ compute_5[(cse_var_1 + 10)] = 0f32
+ compute_5[(cse_var_1 + 11)] = 0f32
+ compute_5[(cse_var_1 + 12)] = 0f32
+ compute_5[(cse_var_1 + 13)] = 0f32
+ compute_5[(cse_var_1 + 14)] = 0f32
+ compute_5[(cse_var_1 + 15)] = 0f32
+ }
+ }
+ for (elem_idx: int32, 0, (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])) {
+ for (i.inner: int32, 0, 2) {
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_2: int32 = ((i.outer.inner*32) + (i.inner*16))
+ compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16))]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_3: int32 = (((i.outer.inner*32) + (i.inner*16)) + 1)
+ compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_4: int32 = (((i.outer.inner*32) + (i.inner*16)) + 2)
+ compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_5: int32 = (((i.outer.inner*32) + (i.inner*16)) + 3)
+ compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_6: int32 = (((i.outer.inner*32) + (i.inner*16)) + 4)
+ compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_7: int32 = (((i.outer.inner*32) + (i.inner*16)) + 5)
+ compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_8: int32 = (((i.outer.inner*32) + (i.inner*16)) + 6)
+ compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_9: int32 = (((i.outer.inner*32) + (i.inner*16)) + 7)
+ compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_10: int32 = (((i.outer.inner*32) + (i.inner*16)) + 8)
+ compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_11: int32 = (((i.outer.inner*32) + (i.inner*16)) + 9)
+ compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_12: int32 = (((i.outer.inner*32) + (i.inner*16)) + 10)
+ compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_13: int32 = (((i.outer.inner*32) + (i.inner*16)) + 11)
+ compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_14: int32 = (((i.outer.inner*32) + (i.inner*16)) + 12)
+ compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_15: int32 = (((i.outer.inner*32) + (i.inner*16)) + 13)
+ compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_16: int32 = (((i.outer.inner*32) + (i.inner*16)) + 14)
+ compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_17: int32 = (((i.outer.inner*32) + (i.inner*16)) + 15)
+ compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
}
}
}
}
- for (i0.inner: int32, 0, 8) {
- let cse_var_132: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*4096) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
- compute[ramp(cse_var_132, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_132, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 128) {
+ let cse_var_18: int32 = ((i0.inner*512) + (i0.outer.i1.outer.fused*16))
+ compute[ramp(cse_var_18, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_18, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -853,7 +549,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 2.694 ms
+ Execution time of this operator: 3.632 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 d46728e8b..5c05a056d 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
@@ -5,12 +5,12 @@
Computation times
=================
-**00:46.715** total execution time for **how_to_tune_with_autotvm** files:
+**00:46.861** total execution time for **how_to_tune_with_autotvm** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:46.678 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:46.825 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.021 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.020 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``) | 00:00.006 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
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 65f778b74..4c6b25efe 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
@@ -1156,8 +1156,8 @@ for this template
TimeoutError
[('tile_f', [-1, 2, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4909501
- No: 9 GFLOPS: 173.43/173.43 result: MeasureResult(costs=(0.0013348717333333334,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8318445682525635, timestamp=1658819800.0006714) [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5072689
- No: 10 GFLOPS: 0.00/173.43 result: Traceback (most recent call last):
+ No: 9 GFLOPS: 122.21/122.21 result: MeasureResult(costs=(0.0018942681607142856,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0902655124664307, timestamp=1658826629.7328916) [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5072689
+ No: 10 GFLOPS: 0.00/122.21 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1280,8 +1280,8 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5092711
- No: 11 GFLOPS: 261.23/261.23 result: MeasureResult(costs=(0.0008862034751381216,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7316110134124756, timestamp=1658819800.9196448) [('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
- No: 12 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+ No: 11 GFLOPS: 261.55/261.55 result: MeasureResult(costs=(0.0008851145303867403,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7998321056365967, timestamp=1658826630.6496062) [('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
+ No: 12 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1404,7 +1404,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 128, 1, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,183542
- No: 13 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+ No: 13 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1527,7 +1527,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 8, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2482196
- No: 14 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+ No: 14 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1650,9 +1650,9 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10306226
- No: 15 GFLOPS: 5.48/261.23 result: MeasureResult(costs=(0.04226059,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.826474666595459, timestamp=1658819805.474508) [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5330964
- No: 16 GFLOPS: 3.36/261.23 result: MeasureResult(costs=(0.06892332475,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.549631118774414, timestamp=1658819806.7095149) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2140058
- No: 17 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+ No: 15 GFLOPS: 5.39/261.55 result: MeasureResult(costs=(0.042985643500000004,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9060659408569336, timestamp=1658826635.3068864) [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5330964
+ No: 16 GFLOPS: 3.35/261.55 result: MeasureResult(costs=(0.06907856475,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.6305437088012695, timestamp=1658826636.5465927) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2140058
+ No: 17 GFLOPS: 0.00/261.55 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/python3.7/concurrent/futures/_base.py", line 435, in result
@@ -1670,8 +1670,8 @@ for this template
TimeoutError
[('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10195251
- No: 18 GFLOPS: 27.15/261.23 result: MeasureResult(costs=(0.0085280948125,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3287360668182373, timestamp=1658819817.7805817) [('tile_f', [-1, 4, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6068603
- No: 19 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+ No: 18 GFLOPS: 28.14/261.55 result: MeasureResult(costs=(0.008227821785714285,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.295943260192871, timestamp=1658826647.5799015) [('tile_f', [-1, 4, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6068603
+ No: 19 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1794,7 +1794,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 16, 4, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6956993
- No: 20 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+ No: 20 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1973,7 +1973,7 @@ and measure running time.
Best config:
[('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
Finish loading 20 records
- Time cost of this operator: 0.001217
+ Time cost of this operator: 0.001272
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 61e3ffde7..09c0cb8a5 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
@@ -329,10 +329,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 311.1 98.734 (1, 2, 10, 10, 3) 2 1 [311.1]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.035 0.963 (1, 6, 10, 10) 1 1 [3.035]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.954 0.303 (1, 1, 10, 10, 3) 1 1 [0.954]
- Total_time - 315.089 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.2 98.732 (1, 2, 10, 10, 3) 2 1 [312.2]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.03 0.958 (1, 6, 10, 10) 1 1 [3.03]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.98 0.31 (1, 1, 10, 10, 3) 1 1 [0.98]
+ Total_time - 316.21 - - - - -
@@ -398,10 +398,10 @@ Timing the tuned program
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 79.688 96.638 (1, 6, 10, 10, 1) 2 1 [79.688]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.802 2.185 (1, 6, 10, 10) 1 1 [1.802]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.97 1.177 (1, 1, 10, 10, 3) 1 1 [0.97]
- Total_time - 82.459 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 79.625 96.391 (1, 6, 10, 10, 1) 2 1 [79.625]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 2.012 2.436 (1, 6, 10, 10) 1 1 [2.012]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.969 1.173 (1, 1, 10, 10, 3) 1 1 [0.969]
+ Total_time - 82.606 - - - - -
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
index 40042035d..5b2950f40 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
@@ -225,7 +225,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
.. code-block:: none
- '/tmp/tmpfld0ithc/images/random'
+ '/tmp/tmpdi_m8g8c/images/random'
@@ -325,8 +325,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmpfld0ithc/images/target contains 8144 images
- /tmp/tmpfld0ithc/images/random contains 5000 images
+ /tmp/tmpdi_m8g8c/images/target contains 8144 images
+ /tmp/tmpdi_m8g8c/images/random contains 5000 images
@@ -501,13 +501,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 55s - loss: 0.2105 - accuracy: 0.9265 - val_loss: 0.1443 - val_accuracy: 0.9577
+ 328/328 - 55s - loss: 0.2380 - accuracy: 0.9218 - val_loss: 0.1577 - val_accuracy: 0.9520
Epoch 2/3
- 328/328 - 52s - loss: 0.0941 - accuracy: 0.9628 - val_loss: 0.1303 - val_accuracy: 0.9622
+ 328/328 - 53s - loss: 0.1005 - accuracy: 0.9638 - val_loss: 0.1300 - val_accuracy: 0.9592
Epoch 3/3
- 328/328 - 52s - loss: 0.0649 - accuracy: 0.9778 - val_loss: 0.1483 - val_accuracy: 0.9551
+ 328/328 - 53s - loss: 0.0670 - accuracy: 0.9748 - val_loss: 0.1283 - val_accuracy: 0.9641
- <keras.callbacks.History object at 0x7f0081213590>
+ <keras.callbacks.History object at 0x7fd5c8c90d50>
@@ -864,7 +864,7 @@ Arduino tutorial for how to do that `on GitHub <https://github.com/guberti/tvm-a
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 5 minutes 26.769 seconds)
+ **Total running time of the script:** ( 5 minutes 14.789 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_train.py:
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 1a1abab8e..1d4e320c1 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
@@ -5,14 +5,14 @@
Computation times
=================
-**06:12.487** total execution time for **how_to_work_with_microtvm** files:
+**06:03.069** total execution time for **how_to_work_with_microtvm** files:
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 05:26.769 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 05:14.789 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:42.434 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:44.804 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.282 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.474 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``) | 00:00.001 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
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 4d4364742..163e634b8 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
@@ -5,14 +5,14 @@
Computation times
=================
-**00:41.507** total execution time for **how_to_work_with_relay** files:
+**00:43.332** total execution time for **how_to_work_with_relay** files:
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:30.038 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:31.637 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:09.706 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:09.944 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.756 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.745 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``) | 00:00.007 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
index d50de846c..7a91aea1a 100644
--- a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
@@ -261,7 +261,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
.. code-block:: none
- <function my_cuda_math_rule at 0x7efff8247710>
+ <function my_cuda_math_rule at 0x7fd5d82b88c0>
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 dc8fb54e9..68a55ec50 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
@@ -5,22 +5,22 @@
Computation times
=================
-**00:04.108** total execution time for **how_to_work_with_schedules** files:
+**00:04.412** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:01.907 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:02.085 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:00.974 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.001 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.532 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.578 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.517 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.561 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.097 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.103 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.040 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.042 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.026 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.027 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.014 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.015 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
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 4028c4b24..c2b56b3dc 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -347,7 +347,7 @@ The importing needs to happen before the tensorized GEMV being executed.
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpdk5__7ic/input0.cc'\nsource_filename = \"/tmp/tmpdk5__7ic/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpavq6yxme/input0.cc'\nsource_filename = \"/tmp/tmpavq6yxme/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
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 5e105b3d3..e35a64864 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
@@ -5,10 +5,10 @@
Computation times
=================
-**00:20.795** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:21.863** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:20.788 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:21.856 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``) | 00:00.006 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``) | 00:00.007 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
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 55048e747..9e244ba9a 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -291,7 +291,7 @@ 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 22.19s!
+ resnet18_v1 inference graph built in 23.92s!
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 0f73de8ac..7c43cfcdf 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -335,7 +335,7 @@ The compilation steps are:
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/relay/build_module.py:411: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 15.70s!
+ yolov3-tiny inference graph built in 16.56s!
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 3f612bdae..3b5c863a3 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
@@ -5,10 +5,10 @@
Computation times
=================
-**01:31.294** total execution time for **topic_vta_tutorials_frontend** files:
+**01:34.098** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:48.830 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:49.570 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:42.464 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:44.528 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
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 6c290e4a7..57a3d0c0f 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
@@ -5,10 +5,10 @@
Computation times
=================
-**00:03.253** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.365** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.850 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.927 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.403 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.438 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
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 98c1ec708..120450019 100644
--- a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:00.731** total execution time for **topic_vta_tutorials** files:
+**00:00.798** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.391 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.421 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.340 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.376 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
index da766647b..de7f0224a 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -205,6 +205,13 @@ trials, we can load the best schedule from the log file and apply it.
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+ .T
+
+
@@ -328,7 +335,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 93.510 ms
+ Execution time of this operator: 94.209 ms
@@ -444,6 +451,11 @@ Expression (TE) language that demonstrates how TVM can optimize computational
operations.
+.. rst-class:: sphx-glr-timing
+
+ **Total running time of the script:** ( 1 minutes 1.315 seconds)
+
+
.. _sphx_glr_download_tutorial_auto_scheduler_matmul_x86.py:
.. only:: html
diff --git a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
index a771f99b5..b03adc61a 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -462,16 +462,16 @@ reduce variance, we take 5 measurements and average them.
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 10.65/10.65 result: MeasureResult(costs=(0.0252159686,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5351924896240234, timestamp=1658818630.013972) [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
- No: 2 GFLOPS: 2.94/10.65 result: MeasureResult(costs=(0.0912649794,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6093063354492188, timestamp=1658818631.6375253) [('tile_y', [-1, 4]), ('tile_x', [-1, 8])],None,32
- No: 3 GFLOPS: 11.93/11.93 result: MeasureResult(costs=(0.022504914199999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5747358798980713, timestamp=1658818632.6790886) [('tile_y', [-1, 64]), ('tile_x', [-1, 32])],None,56
- No: 4 GFLOPS: 1.87/11.93 result: MeasureResult(costs=(0.14385245480000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.4205527305603027, timestamp=1658818635.655591) [('tile_y', [-1, 1]), ('tile_x', [-1, 4])],None,20
- No: 5 GFLOPS: 3.69/11.93 result: MeasureResult(costs=(0.0727963934,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3022141456604004, timestamp=1658818637.0808384) [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
- No: 6 GFLOPS: 1.83/11.93 result: MeasureResult(costs=(0.1465060762,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.4773406982421875, timestamp=1658818640.1136973) [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
- No: 7 GFLOPS: 0.87/11.93 result: MeasureResult(costs=(0.3101616356,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.085535049438477, timestamp=1658818645.2464063) [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
- No: 8 GFLOPS: 10.89/11.93 result: MeasureResult(costs=(0.0246409192,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5361745357513428, timestamp=1658818645.8030849) [('tile_y', [-1, 4]), ('tile_x', [-1, 64])],None,62
- No: 9 GFLOPS: 1.92/11.93 result: MeasureResult(costs=(0.13983667959999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3361587524414062, timestamp=1658818648.2591856) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
- No: 10 GFLOPS: 2.47/11.93 result: MeasureResult(costs=(0.10847894360000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8392126560211182, timestamp=1658818650.1572676) [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
+ No: 1 GFLOPS: 8.93/8.93 result: MeasureResult(costs=(0.030073501199999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6201331615447998, timestamp=1658825403.434856) [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
+ No: 2 GFLOPS: 2.67/8.93 result: MeasureResult(costs=(0.10060184860000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7668182849884033, timestamp=1658825405.7655199) [('tile_y', [-1, 4]), ('tile_x', [-1, 8])],None,32
+ No: 3 GFLOPS: 11.69/11.69 result: MeasureResult(costs=(0.022964935,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5593230724334717, timestamp=1658825406.851242) [('tile_y', [-1, 64]), ('tile_x', [-1, 32])],None,56
+ No: 4 GFLOPS: 1.57/11.69 result: MeasureResult(costs=(0.171481672,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.853118896484375, timestamp=1658825409.7603683) [('tile_y', [-1, 1]), ('tile_x', [-1, 4])],None,20
+ No: 5 GFLOPS: 3.56/11.69 result: MeasureResult(costs=(0.0754066312,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3550024032592773, timestamp=1658825411.243578) [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
+ No: 6 GFLOPS: 1.80/11.69 result: MeasureResult(costs=(0.1489293424,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5476973056793213, timestamp=1658825413.8359394) [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
+ No: 7 GFLOPS: 0.85/11.69 result: MeasureResult(costs=(0.315868708,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.182083368301392, timestamp=1658825419.6049476) [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
+ No: 8 GFLOPS: 9.51/11.69 result: MeasureResult(costs=(0.0282242828,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6093072891235352, timestamp=1658825420.2233741) [('tile_y', [-1, 4]), ('tile_x', [-1, 64])],None,62
+ No: 9 GFLOPS: 1.83/11.69 result: MeasureResult(costs=(0.1467900244,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.4650285243988037, timestamp=1658825422.8092453) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
+ No: 10 GFLOPS: 2.20/11.69 result: MeasureResult(costs=(0.1222558732,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.064823627471924, timestamp=1658825424.932038) [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index c68f2b634..65b8e4ca9 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -327,7 +327,7 @@ standard deviation.
.. code-block:: none
- {'mean': 490.1264289199389, 'median': 489.95082740002545, 'std': 0.772977499482236}
+ {'mean': 497.18471650999163, 'median': 497.1574910000072, 'std': 0.47640301669203194}
@@ -563,31 +563,31 @@ the tuning data to.
/workspace/python/tvm/driver/build_module.py:268: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
-
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 17.55/ 17.55 GFLOPS | Progress: (4/20) | 6.27 s
[Task 1/25] Current/Best: 6.16/ 17.55 GFLOPS | Progress: (8/20) | 9.18 s
[Task 1/25] Current/Best: 11.58/ 22.87 GFLOPS | Progress: (12/20) | 11.62 s
[Task 1/25] Current/Best: 16.82/ 22.87 GFLOPS | Progress: (16/20) | 13.30 s
[Task 1/25] Current/Best: 11.64/ 23.89 GFLOPS | Progress: (20/20) | 15.03 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.36/ 13.20 GFLOPS | Progress: (4/20) | 3.72 s
[Task 2/25] Current/Best: 14.05/ 18.30 GFLOPS | Progress: (8/20) | 5.02 s
[Task 2/25] Current/Best: 21.20/ 21.20 GFLOPS | Progress: (12/20) | 6.37 s
[Task 2/25] Current/Best: 12.14/ 21.20 GFLOPS | Progress: (16/20) | 7.69 s
[Task 2/25] Current/Best: 19.62/ 21.20 GFLOPS | Progress: (20/20) | 9.28 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 1.63/ 10.60 GFLOPS | Progress: (4/20) | 5.85 s
[Task 3/25] Current/Best: 15.61/ 16.89 GFLOPS | Progress: (8/20) | 7.78 s
[Task 3/25] Current/Best: 14.94/ 16.89 GFLOPS | Progress: (12/20) | 9.53 s
[Task 3/25] Current/Best: 7.22/ 23.82 GFLOPS | Progress: (16/20) | 11.44 s
[Task 3/25] Current/Best: 12.73/ 23.82 GFLOPS | Progress: (20/20) | 16.03 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 9.47/ 18.19 GFLOPS | Progress: (4/20) | 2.41 s
[Task 4/25] Current/Best: 6.87/ 18.19 GFLOPS | Progress: (8/20) | 7.12 s
[Task 4/25] Current/Best: 21.73/ 21.73 GFLOPS | Progress: (12/20) | 12.11 s
[Task 4/25] Current/Best: 16.75/ 21.73 GFLOPS | Progress: (16/20) | 14.52 s
[Task 4/25] Current/Best: 13.38/ 21.73 GFLOPS | Progress: (20/20) | 16.58 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 9.91/ 10.51 GFLOPS | Progress: (4/20) | 2.55 s
[Task 5/25] Current/Best: 11.95/ 13.04 GFLOPS | Progress: (8/20) | 4.58 s
[Task 5/25] Current/Best: 11.66/ 18.06 GFLOPS | Progress: (12/20) | 7.65 s
[Task 5/25] Current/Best: 11.91/ 22.77 GFLOPS | Progress: (16/20) | 9.06 s
[Task 5/25] Current/Best: 12.06/ 22.77 GFLOPS | Progress: (20/20) | 10.96 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 12.25/ 20.78 GFLOPS | Progress: (4/20) | 4.07 s
[Task 6/25] Current/Best: 18.96/ 20.78 GFLOPS | Progress: (8/20) | 5.82 s
[Task 6/25] Current/Best: 13.33/ 20.78 GFLOPS | Progress: (12/20) | 7.75 s
[Task 6/25] Current/Best: 20.09/ 20.78 GFLOPS | Progress: (16/20) | 9.99 s
[Task 6/25] Current/Best: 3.74/ 20.78 GFLOPS | Progress: (20/20) | 12.50 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 11.12/ 13.03 GFLOPS | Progress: (4/20) | 3.59 s
[Task 7/25] Current/Best: 20.28/ 21.20 GFLOPS | Progress: (8/20) | 5.08 s
[Task 7/25] Current/Best: 15.85/ 21.20 GFLOPS | Progress: (12/20) | 6.97 s
[Task 7/25] Current/Best: 12.30/ 21.20 GFLOPS | Progress: (16/20) | 8.99 s
[Task 7/25] Current/Best: 6.40/ 21.78 GFLOPS | Progress: (20/20) | 11.43 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 10.28/ 14.25 GFLOPS | Progress: (4/20) | 2.87 s
[Task 8/25] Current/Best: 9.69/ 14.25 GFLOPS | Progress: (8/20) | 7.94 s
[Task 8/25] Current/Best: 12.69/ 14.25 GFLOPS | Progress: (12/20) | 14.44 s
[Task 8/25] Current/Best: 18.77/ 18.77 GFLOPS | Progress: (16/20) | 16.50 s
[Task 8/25] Current/Best: 19.91/ 19.91 GFLOPS | Progress: (20/20) | 23.63 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 14.40/ 15.16 GFLOPS | Progress: (4/20) | 11.94 s
[Task 9/25] Current/Best: 23.55/ 23.55 GFLOPS | Progress: (8/20) | 13.72 s
[Task 9/25] Current/Best: 8.27/ 23.55 GFLOPS | Progress: (12/20) | 16.27 s
[Task 9/25] Current/Best: 18.03/ 23.55 GFLOPS | Progress: (16/20) | 19.15 s
[Task 9/25] Current/Best: 9.26/ 23.55 GFLOPS | Progress: (20/20) | 27.69 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 18.24/ 18.24 GFLOPS | Progress: (4/20) | 2.56 s
[Task 10/25] Current/Best: 15.58/ 18.24 GFLOPS | Progress: (8/20) | 4.21 s
[Task 10/25] Current/Best: 12.89/ 19.12 GFLOPS | Progress: (12/20) | 5.74 s
[Task 10/25] Current/Best: 19.16/ 20.42 GFLOPS | Progress: (16/20) | 6.86 s
[Task 10/25] Current/Best: 8.86/ 20.42 GFLOPS | Progress: (20/20
) | 8.41 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 12.35/ 18.22 GFLOPS | Progress: (4/20) | 3.29 s
[Task 11/25] Current/Best: 17.02/ 18.22 GFLOPS | Progress: (8/20) | 6.09 s
[Task 11/25] Current/Best: 18.25/ 18.25 GFLOPS | Progress: (12/20) | 8.16 s
[Task 11/25] Current/Best: 13.42/ 21.25 GFLOPS | Progress: (16/20) | 11.11 s
[Task 11/25] Current/Best: 19.43/ 21.63 GFLOPS | Progress: (20/20) | 13.19 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 7.86/ 17.97 GFLOPS | Progress: (4/20) | 5.76 s
[Task 12/25] Current/Best: 5.31/ 17.97 GFLOPS | Progress: (8/20) | 9.71 s
[Task 12/25] Current/Best: 18.98/ 18.98 GFLOPS | Progress: (12/20) | 11.69 s
[Task 12/25] Current/Best: 15.50/ 18.98 GFLOPS | Progress: (16/20) | 14.57 s
[Task 12/25] Current/Best: 15.14/ 19.11 GFLOPS | Progress: (20/20) | 16.47 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 8.70/ 17.30 GFLOPS | Progress: (4/20) | 3.70 s
[Task 13/25] Current/Best: 16.16/ 21.14 GFLOPS | Progress: (8/20) | 6.29 s
[Task 13/25] Current/Best: 19.57/ 21.37 GFLOPS | Progress: (12/20) | 9.31 s
[Task 13/25] Current/Best: 12.30/ 21.37 GFLOPS | Progress: (16/20) | 12.74 s
[Task 13/25] Current/Best: 19.03/ 21.37 GFLOPS | Progress: (20/20) | 15.07 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 13.60/ 13.60 GFLOPS | Progress: (4/20) | 3.39 s
[Task 14/25] Current/Best: 6.11/ 13.60 GFLOPS | Progress: (8/20) | 5.55 s
[Task 14/25] Current/Best: 20.83/ 20.83 GFLOPS | Progress: (12/20) | 8.24 s
[Task 14/25] Current/Best: 17.49/ 20.83 GFLOPS | Progress: (16/20) | 9.89 s Done.
-
[Task 14/25] Current/Best: 17.38/ 20.83 GFLOPS | Progress: (20/20) | 11.69 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 16.17/ 17.69 GFLOPS | Progress: (4/20) | 2.68 s
[Task 15/25] Current/Best: 14.47/ 18.17 GFLOPS | Progress: (8/20) | 4.02 s
[Task 15/25] Current/Best: 10.38/ 22.36 GFLOPS | Progress: (12/20) | 6.25 s
[Task 15/25] Current/Best: 20.43/ 22.36 GFLOPS | Progress: (16/20) | 9.64 s
[Task 15/25] Current/Best: 9.72/ 22.36 GFLOPS | Progress: (20/20) | 10.65 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 19.47/ 19.47 GFLOPS | Progress: (4/20) | 2.91 s
[Task 16/25] Current/Best: 3.04/ 19.47 GFLOPS | Progress: (8/20) | 4.53 s
[Task 16/25] Current/Best: 19.76/ 19.76 GFLOPS | Progress: (12/20) | 5.74 s
[Task 16/25] Current/Best: 18.14/ 19.76 GFLOPS | Progress: (16/20) |
7.12 s
[Task 16/25] Current/Best: 9.92/ 22.41 GFLOPS | Progress: (20/20) | 9.25 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 12.94/ 18.69 GFLOPS | Progress: (4/20) | 4.80 s
[Task 17/25] Current/Best: 14.50/ 23.39 GFLOPS | Progress: (8/20) | 7.68 s
[Task 17/25] Current/Best: 16.88/ 23.39 GFLOPS | Progress: (12/20) | 9.73 s
[Task 17/25] Current/Best: 16.78/ 23.39 GFLOPS | Progress: (16/20) | 11.94 s
[Task 17/25] Current/Best: 10.05/ 23.39 GFLOPS | Progress: (20/20) | 14.07 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 11.38/ 17.80 GFLOPS | Progress: (4/20) | 3.79 s
[Task 18/25] Current/Best: 10.55/ 20.09 GFLOPS | Progress: (8/20) | 7.44 s
[Task 18/25] Current/Best: 18.74/ 20.09 GFLOPS | Progress: (12/20) | 9.38 s
[Task 18/25] Current/Best: 10.05/ 20.09 GFLOPS | Progress: (16/20) | 13.25 s
[Task 18/25] Current/Best: 20.68/ 20.68 GFLOPS | Progress: (20/20) | 14.76 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 7.21/ 20.41 GFLOPS | Progress: (4/20) | 6.00 s
[Task 19/25] Current/Best: 2.61/ 20.41 GFLOPS | Progress: (8/20) | 9.37 s
[Task 19/25] Current/Best: 20.63/ 21.90 GFLOPS | Progress: (12/20) | 12.38 s
[Task 19/25] Current/Best: 14.24/ 21.90 GFLOPS | Progress: (16/20) | 15.48 s
[Task 19/25] Current/Best: 2.70/ 23.54 GFLOPS | Progress: (20/20) | 18.32 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 8.45/ 15.32 GFLOPS | Progress: (4/20) | 3.31 s Done.
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 17.43/ 17.43 GFLOPS | Progress: (4/20) | 6.41 s
[Task 1/25] Current/Best: 6.15/ 17.43 GFLOPS | Progress: (8/20) | 9.45 s
[Task 1/25] Current/Best: 11.50/ 22.69 GFLOPS | Progress: (12/20) | 11.99 s
[Task 1/25] Current/Best: 16.69/ 22.74 GFLOPS | Progress: (16/20) | 13.68 s
[Task 1/25] Current/Best: 11.57/ 23.82 GFLOPS | Progress: (20/20) | 15.44 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.22/ 12.94 GFLOPS | Progress: (4/20) | 3.86 s
[Task 2/25] Current/Best: 14.09/ 18.44 GFLOPS | Progress: (8/20) | 5.16 s
[Task 2/25] Current/Best: 21.05/ 21.05 GFLOPS | Progress: (12/20) | 6.50 s
[Task 2/25] Current/Best: 11.97/ 21.05 GFLOPS | Progress: (16/20) | 7.78 s
[Task 2/25] Current/Best: 20.03/ 21.05 GFLOPS | Progress: (20/20) | 9.43 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 1.62/ 10.53 GFLOPS | Progress: (4/20) | 5.92 s
[Task 3/25] Current/Best: 15.48/ 16.82 GFLOPS | Progress: (8/20) | 7.86 s
[Task 3/25] Current/Best: 14.81/ 16.82 GFLOPS | Progress: (12/20) | 9.59 s
[Task 3/25] Current/Best: 7.20/ 23.69 GFLOPS | Progress: (16/20) | 11.54 s
[Task 3/25] Current/Best: 12.58/ 23.69 GFLOPS | Progress: (20/20) | 16.15 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 9.57/ 20.34 GFLOPS | Progress: (4/20) | 2.44 s
[Task 4/25] Current/Best: 6.73/ 20.34 GFLOPS | Progress: (8/20) | 7.27 s
[Task 4/25] Current/Best: 21.38/ 21.38 GFLOPS | Progress: (12/20) | 12.24 s
[Task 4/25] Current/Best: 16.57/ 21.38 GFLOPS | Progress: (16/20) | 14.69 s
[Task 4/25] Current/Best: 13.23/ 21.38 GFLOPS | Progress: (20/20) | 16.72 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 9.78/ 10.13 GFLOPS | Progress: (4/20) | 2.66 s
[Task 5/25] Current/Best: 11.70/ 12.96 GFLOPS | Progress: (8/20) | 4.74 s
[Task 5/25] Current/Best: 10.47/ 18.25 GFLOPS | Progress: (12/20) | 8.03 s
[Task 5/25] Current/Best: 11.63/ 22.70 GFLOPS | Progress: (16/20) | 9.50 s
[Task 5/25] Current/Best: 12.08/ 22.70 GFLOPS | Progress: (20/20) | 11.44 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 12.15/ 20.69 GFLOPS | Progress: (4/20) | 4.19 s
[Task 6/25] Current/Best: 18.90/ 20.69 GFLOPS | Progress: (8/20) | 5.97 s
[Task 6/25] Current/Best: 13.31/ 20.69 GFLOPS | Progress: (12/20) | 7.93 s
[Task 6/25] Current/Best: 19.90/ 20.69 GFLOPS | Progress: (16/20) | 10.22 s
[Task 6/25] Current/Best: 3.75/ 20.69 GFLOPS | Progress: (20/20) | 12.78 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 9.96/ 12.31 GFLOPS | Progress: (4/20) | 3.75 s
[Task 7/25] Current/Best: 20.27/ 20.99 GFLOPS | Progress: (8/20) | 5.28 s
[Task 7/25] Current/Best: 16.03/ 20.99 GFLOPS | Progress: (12/20) | 7.20 s
[Task 7/25] Current/Best: 12.26/ 20.99 GFLOPS | Progress: (16/20) | 9.29 s
[Task 7/25] Current/Best: 6.32/ 21.66 GFLOPS | Progress: (20/20) | 11.75 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 9.83/ 14.37 GFLOPS | Progress: (4/20) | 2.98 s
[Task 8/25] Current/Best: 9.77/ 14.37 GFLOPS | Progress: (8/20) | 8.25 s
[Task 8/25] Current/Best: 12.34/ 14.37 GFLOPS | Progress: (12/20) | 14.94 s
[Task 8/25] Current/Best: 18.95/ 18.95 GFLOPS | Progress: (16/20) | 17.03 s
[Task 8/25] Current/Best: 19.75/ 19.75 GFLOPS | Progress: (20/20) | 24.28 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 14.10/ 15.75 GFLOPS | Progress: (4/20) | 12.02 s
[Task 9/25] Current/Best: 23.52/ 23.52 GFLOPS | Progress: (8/20) | 13.86 s
[Task 9/25] Current/Best: 8.23/ 23.52 GFLOPS | Progress: (12/20) | 16.45 s
[Task 9/25] Current/Best: 17.76/ 23.52 GFLOPS | Progress: (16/20) | 19.29 s
[Task 9/25] Current/Best: 9.10/ 23.52 GFLOPS | Progress: (20/20) | 28.26 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 18.02/ 18.02 GFLOPS | Progress: (4/20) | 2.62 s
[Task 10/25] Current/Best: 15.62/ 18.02 GFLOPS | Progress: (8/20) | 4.28 s
[Task 10/25] Current/Best: 12.52/ 19.15 GFLOPS | Progress: (12/20) | 5.86 s
[Task 10/25] Current/Best: 19.07/ 20.62 GFLOPS | Progress: (16/20) | 6.99 s
[Task 10/25] Current/Best: 8.94/ 20.62 GFLOPS | Progress: (20/20
) | 8.55 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 12.26/ 18.13 GFLOPS | Progress: (4/20) | 3.46 s
[Task 11/25] Current/Best: 16.82/ 18.13 GFLOPS | Progress: (8/20) | 6.34 s
[Task 11/25] Current/Best: 17.38/ 18.13 GFLOPS | Progress: (12/20) | 8.46 s
[Task 11/25] Current/Best: 13.25/ 20.98 GFLOPS | Progress: (16/20) | 11.48 s
[Task 11/25] Current/Best: 19.36/ 21.36 GFLOPS | Progress: (20/20) | 13.64 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 7.78/ 18.28 GFLOPS | Progress: (4/20) | 5.92 s
[Task 12/25] Current/Best: 5.20/ 18.28 GFLOPS | Progress: (8/20) | 9.94 s
[Task 12/25] Current/Best: 19.14/ 19.14 GFLOPS | Progress: (12/20) | 11.94 s
[Task 12/25] Current/Best: 15.27/ 19.14 GFLOPS | Progress: (16/20) | 14.93 s
[Task 12/25] Current/Best: 15.20/ 19.14 GFLOPS | Progress: (20/20) | 16.89 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 8.95/ 17.19 GFLOPS | Progress: (4/20) | 3.84 s
[Task 13/25] Current/Best: 15.73/ 21.03 GFLOPS | Progress: (8/20) | 6.49 s
[Task 13/25] Current/Best: 19.50/ 21.22 GFLOPS | Progress: (12/20) | 9.56 s
[Task 13/25] Current/Best: 12.23/ 21.22 GFLOPS | Progress: (16/20) | 13.05 s
[Task 13/25] Current/Best: 18.47/ 21.22 GFLOPS | Progress: (20/20) | 15.42 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 13.31/ 13.33 GFLOPS | Progress: (4/20) | 3.50 s
[Task 14/25] Current/Best: 6.11/ 13.33 GFLOPS | Progress: (8/20) | 5.74 s
[Task 14/25] Current/Best: 20.14/ 20.14 GFLOPS | Progress: (12/20) | 8.44 s
[Task 14/25] Current/Best: 16.82/ 20.14 GFLOPS | Progress: (16/20) | 10.11 s Done.
+
[Task 14/25] Current/Best: 17.25/ 20.14 GFLOPS | Progress: (20/20) | 11.91 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 15.99/ 17.58 GFLOPS | Progress: (4/20) | 2.79 s
[Task 15/25] Current/Best: 14.39/ 17.76 GFLOPS | Progress: (8/20) | 4.14 s
[Task 15/25] Current/Best: 10.37/ 22.27 GFLOPS | Progress: (12/20) | 6.52 s
[Task 15/25] Current/Best: 20.35/ 22.27 GFLOPS | Progress: (16/20) | 9.74 s
[Task 15/25] Current/Best: 9.67/ 22.27 GFLOPS | Progress: (20/20) | 10.77 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 20.36/ 20.36 GFLOPS | Progress: (4/20) | 3.07 s
[Task 16/25] Current/Best: 3.04/ 20.36 GFLOPS | Progress: (8/20) | 4.69 s
[Task 16/25] Current/Best: 19.49/ 20.36 GFLOPS | Progress: (12/20) | 5.91 s
[Task 16/25] Current/Best: 17.56/ 20.36 GFLOPS | Progress: (16/20) |
7.29 s
[Task 16/25] Current/Best: 9.86/ 22.28 GFLOPS | Progress: (20/20) | 9.49 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 13.34/ 18.94 GFLOPS | Progress: (4/20) | 4.89 s
[Task 17/25] Current/Best: 14.40/ 23.01 GFLOPS | Progress: (8/20) | 7.89 s
[Task 17/25] Current/Best: 16.89/ 23.01 GFLOPS | Progress: (12/20) | 9.96 s
[Task 17/25] Current/Best: 16.49/ 23.01 GFLOPS | Progress: (16/20) | 12.21 s
[Task 17/25] Current/Best: 10.02/ 23.01 GFLOPS | Progress: (20/20) | 14.40 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 11.28/ 18.01 GFLOPS | Progress: (4/20) | 3.90 s
[Task 18/25] Current/Best: 10.58/ 18.18 GFLOPS | Progress: (8/20) | 7.68 s
[Task 18/25] Current/Best: 19.16/ 19.16 GFLOPS | Progress: (12/20) | 9.63 s
[Task 18/25] Current/Best: 9.92/ 19.16 GFLOPS | Progress: (16/20) | 13.60 s
[Task 18/25] Current/Best: 20.66/ 20.66 GFLOPS | Progress: (20/20) | 15.15 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 6.87/ 20.27 GFLOPS | Progress: (4/20) | 6.28 s
[Task 19/25] Current/Best: 2.61/ 20.27 GFLOPS | Progress: (8/20) | 9.64 s
[Task 19/25] Current/Best: 19.35/ 21.31 GFLOPS | Progress: (12/20) | 12.64 s
[Task 19/25] Current/Best: 15.39/ 21.61 GFLOPS | Progress: (16/20) | 15.64 s
[Task 19/25] Current/Best: 2.70/ 23.09 GFLOPS | Progress: (20/20) | 18.44 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 8.57/ 14.58 GFLOPS | Progress: (4/20) | 3.45 s Done.
Done.
-
[Task 20/25] Current/Best: 9.77/ 15.32 GFLOPS | Progress: (8/20) | 6.71 s
[Task 20/25] Current/Best: 2.32/ 16.66 GFLOPS | Progress: (12/20) | 10.61 s
[Task 20/25] Current/Best: 12.35/ 16.66 GFLOPS | Progress: (16/20) | 14.54 s
[Task 20/25] Current/Best: 12.47/ 22.36 GFLOPS | Progress: (20/20) | 16.67 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 6.40/ 17.70 GFLOPS | Progress: (4/20) | 3.26 s
[Task 21/25] Current/Best: 14.69/ 17.70 GFLOPS | Progress: (8/20) | 4.81 s
[Task 21/25] Current/Best: 1.61/ 17.70 GFLOPS | Progress: (12/20) | 6.92 s
[Task 21/25] Current/Best: 18.10/ 18.10 GFLOPS | Progress: (16/20) | 10.41 s
[Task 21/25] Current/Best: 4.48/ 18.10 GFLOPS | Progress: (20/20) | 17.64 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 2.71/ 16.98 GFLOPS | Progress: (4/20
) | 2.66 s
[Task 22/25] Current/Best: 8.60/ 22.08 GFLOPS | Progress: (8/20) | 4.72 s
[Task 22/25] Current/Best: 20.13/ 22.08 GFLOPS | Progress: (12/20) | 7.11 s
[Task 22/25] Current/Best: 15.55/ 22.08 GFLOPS | Progress: (16/20) | 9.26 s
[Task 22/25] Current/Best: 14.17/ 22.08 GFLOPS | Progress: (20/20) | 10.98 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 17.63/ 20.71 GFLOPS | Progress: (4/20) | 3.23 s
[Task 23/25] Current/Best: 14.36/ 20.71 GFLOPS | Progress: (8/20) | 6.52 s
[Task 23/25] Current/Best: 21.08/ 21.71 GFLOPS | Progress: (12/20) | 8.36 s
[Task 23/25] Current/Best: 6.41/ 21.71 GFLOPS | Progress: (16/20) | 15.22 s
[Task 23/25] Current/Best: 7.76/ 21.71 GFLOPS | Progress: (20/20) | 19.42 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.39/ 8.39 GFLOPS | Progress: (4/20) | 11.80 s
[Task 24/25] Current/Best: 2.16/ 8.39 GFLOPS | Progress: (8/20) | 22.82 s
[Task 24/25] Current/Best: 4.44/ 8.39 GFLOPS | Progress: (12/20) | 34.36 s Done.
+
[Task 20/25] Current/Best: 10.42/ 14.58 GFLOPS | Progress: (8/20) | 6.90 s
[Task 20/25] Current/Best: 2.32/ 16.66 GFLOPS | Progress: (12/20) | 10.92 s
[Task 20/25] Current/Best: 12.43/ 16.66 GFLOPS | Progress: (16/20) | 14.94 s
[Task 20/25] Current/Best: 13.61/ 21.75 GFLOPS | Progress: (20/20) | 17.05 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 6.39/ 17.49 GFLOPS | Progress: (4/20) | 3.38 s
[Task 21/25] Current/Best: 14.40/ 17.49 GFLOPS | Progress: (8/20) | 5.04 s
[Task 21/25] Current/Best: 1.61/ 17.49 GFLOPS | Progress: (12/20) | 7.23 s
[Task 21/25] Current/Best: 17.73/ 17.73 GFLOPS | Progress: (16/20) | 10.83 s
[Task 21/25] Current/Best: 4.47/ 17.73 GFLOPS | Progress: (20/20) | 18.29 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 2.70/ 16.99 GFLOPS | Progress: (4/20
) | 2.77 s
[Task 22/25] Current/Best: 9.17/ 22.11 GFLOPS | Progress: (8/20) | 4.77 s
[Task 22/25] Current/Best: 19.88/ 22.11 GFLOPS | Progress: (12/20) | 7.19 s
[Task 22/25] Current/Best: 14.86/ 22.11 GFLOPS | Progress: (16/20) | 9.38 s
[Task 22/25] Current/Best: 14.73/ 22.11 GFLOPS | Progress: (20/20) | 11.14 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 17.40/ 20.18 GFLOPS | Progress: (4/20) | 3.33 s
[Task 23/25] Current/Best: 15.80/ 20.18 GFLOPS | Progress: (8/20) | 6.65 s
[Task 23/25] Current/Best: 20.68/ 21.32 GFLOPS | Progress: (12/20) | 8.55 s
[Task 23/25] Current/Best: 6.25/ 21.32 GFLOPS | Progress: (16/20) | 15.83 s
[Task 23/25] Current/Best: 7.58/ 21.32 GFLOPS | Progress: (20/20) | 20.16 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.62/ 8.62 GFLOPS | Progress: (4/20) | 11.86 s
[Task 24/25] Current/Best: 3.24/ 8.62 GFLOPS | Progress: (8/20) | 23.15 s
[Task 24/25] Current/Best: 4.31/ 8.62 GFLOPS | Progress: (12/20) | 33.94 s Done.
Done.
-
[Task 24/25] Current/Best: 6.25/ 8.80 GFLOPS | Progress: (16/20) | 40.08 s
[Task 24/25] Current/Best: 3.38/ 8.80 GFLOPS | Progress: (20/20) | 45.91 s Done.
-
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 25/25] Current/Best: 1.55/ 2.73 GFLOPS | Progress: (4/20) | 11.59 s
[Task 25/25] Current/Best: 6.09/ 8.53 GFLOPS | Progress: (8/20) | 22.83 s
[Task 25/25] Current/Best: 6.14/ 8.53 GFLOPS | Progress: (12/20) | 34.11 s
[Task 25/25] Current/Best: 5.82/ 9.07 GFLOPS | Progress: (16/20) | 35.93 s
[Task 25/25] Current/Best: 2.87/ 9.35 GFLOPS | Progress: (20/20) | 46.64 s
+
[Task 24/25] Current/Best: 7.19/ 8.95 GFLOPS | Progress: (16/20) | 39.86 s
[Task 24/25] Current/Best: 3.30/ 8.99 GFLOPS | Progress: (20/20) | 45.97 s Done.
+
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 25/25] Current/Best: 1.55/ 2.96 GFLOPS | Progress: (4/20) | 11.65 s
[Task 25/25] Current/Best: 5.71/ 7.71 GFLOPS | Progress: (8/20) | 22.97 s
[Task 25/25] Current/Best: 5.91/ 7.71 GFLOPS | Progress: (12/20) | 34.31 s
[Task 25/25] Current/Best: 5.75/ 9.54 GFLOPS | Progress: (16/20) | 36.07 s
[Task 25/25] Current/Best: 2.91/ 9.54 GFLOPS | Progress: (20/20) | 46.76 s
@@ -748,8 +748,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 407.95316013000047, 'median': 407.83968870000535, 'std': 0.3405536191071367}
- unoptimized: {'mean': 490.1264289199389, 'median': 489.95082740002545, 'std': 0.772977499482236}
+ optimized: {'mean': 415.5701937199865, 'median': 414.8568008499751, 'std': 1.9431914272223534}
+ unoptimized: {'mean': 497.18471650999163, 'median': 497.1574910000072, 'std': 0.47640301669203194}
@@ -772,7 +772,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 10 minutes 21.388 seconds)
+ **Total running time of the script:** ( 10 minutes 36.705 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 14629f729..aad39ef63 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -282,7 +282,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.194e-07 secs/op
+ 1.261e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 42e7153b5..35325fd04 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -263,7 +263,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0xd73ff70)), stage(b, placeholder(b, 0x10f393d0)), 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 [...]
+ [stage(a, placeholder(a, 0x2176a980)), stage(b, placeholder(b, 0x208d2c80)), 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 [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index f97a6ec62..a5c52c304 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,26 +5,26 @@
Computation times
=================
-**13:01.926** total execution time for **tutorial** files:
+**13:36.954** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 10:21.388 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 10:36.705 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 00:58.368 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:01.492 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 00:47.408 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:01.315 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:29.616 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:30.597 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:23.555 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:25.166 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.744 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.768 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.691 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.715 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.149 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.188 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``) | 00:00.005 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``) | 00:00.004 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``) | 00:00.001 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index 8dfa1a33f..ab8f5419d 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -301,8 +301,8 @@ helper function to run a profile of the TVM generated code.
.. code-block:: none
- Numpy running time: 0.000008
- naive: 0.000007
+ Numpy running time: 0.000007
+ naive: 0.000009
@@ -403,7 +403,7 @@ compile and run this new schedule with the parallel operation applied:
/workspace/python/tvm/driver/build_module.py:268: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- parallel: 0.000008
+ parallel: 0.000006
@@ -512,10 +512,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 8.109930004138732e-06 1.0
- naive 6.815300000000001e-06 0.8403648362590005
- parallel 7.705399999999999e-06 0.9501191744031958
- vector 2.4612400000000002e-05 3.0348474015730815
+ numpy 7.1167300029628674e-06 1.0
+ naive 8.7248e-06 1.2259563024545899
+ parallel 6.082900000000001e-06 0.8547324399643574
+ vector 2.4669900000000002e-05 3.466465636567542
@@ -936,7 +936,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.018570
+ Numpy running time: 0.019317
@@ -996,7 +996,7 @@ optimizations.
/workspace/python/tvm/driver/build_module.py:268: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- none: 3.237281
+ none: 3.430042
@@ -1101,7 +1101,7 @@ schedule.
/workspace/python/tvm/driver/build_module.py:268: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- blocking: 0.289066
+ blocking: 0.310437
@@ -1199,7 +1199,7 @@ already cache friendly from our previous optimizations.
/workspace/python/tvm/driver/build_module.py:268: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- vectorization: 0.331122
+ vectorization: 0.343972
@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], []),
@@ -1275,7 +1275,7 @@ more cache friendly.
/workspace/python/tvm/driver/build_module.py:268: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- loop permutation: 0.116312
+ loop permutation: 0.120048
@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], []),
@@ -1376,7 +1376,7 @@ optimized schedule.
/workspace/python/tvm/driver/build_module.py:268: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- array packing: 0.110486
+ array packing: 0.110687
@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], []),
@@ -1471,7 +1471,7 @@ to `C` when all the block results are ready.
/workspace/python/tvm/driver/build_module.py:268: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- block caching: 0.110768
+ block caching: 0.110726
@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], []),
@@ -1559,7 +1559,7 @@ of thread-level parallelization.
/workspace/python/tvm/driver/build_module.py:268: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- parallelization: 0.144061
+ parallelization: 0.144799
@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], []),
@@ -1640,13 +1640,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.2372814725 1.0
- blocking 0.2890661727 0.08929287587611831
- vectorization 0.33112216400000005 0.10228402034633399
- loop permutation 0.11631215179999999 0.035928958537602106
- array packing 0.110486042 0.03412926646587726
- block caching 0.1107683951 0.03421648566581354
- parallelization 0.14406051150000002 0.044500459019014144
+ none 3.4300416061999996 1.0
+ blocking 0.3104366301 0.09050520831551073
+ vectorization 0.3439716159 0.10028205351161086
+ loop permutation 0.1200484297 0.034999117644230755
+ array packing 0.1106868022 0.032269813287374466
+ block caching 0.11072593580000001 0.032281222361809386
+ parallelization 0.1447985788 0.04221481702678712
@@ -1686,6 +1686,11 @@ 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 1.492 seconds)
+
+
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
.. only:: html
diff --git a/docs/commit_hash b/docs/commit_hash
index 980d152b4..f3188d64b 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-9bef7de9f0b6611b5a1b6fef3fef722334d8c9f0
+ea6ea42757f275573391eb3ff67034b2749948ae
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index 460d20035..3741105ec 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -569,7 +569,7 @@ class:['truck 0.9266'] left:471 top:83 right:689 bottom:169
class:['bicycle 0.9984'] left:111 top:113 right:577 bottom:447
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 1.933 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 3.724 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-darknet-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7716f96385bd5abb6e822041e285be54/from_darknet.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_darknet.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 31643d1fd..d187acc74 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -422,7 +422,7 @@ to download the full example code</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"x"</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">x</span><span class="o">.</span><span class="n">shape</span></a><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip52e1b03f-7187-44e0-8eaf-9940e8260f8f from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip8f3694ae-a876-4feb-a97d-0081dca49141 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_oneflow.html b/docs/how_to/compile_models/from_oneflow.html
index e6fdec4a8..ee989eda9 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -427,14 +427,13 @@ python3 -m pip install -f https://release.oneflow.info <span class="nv">oneflow<
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
0%| | 0.00/41.5M [00:00<?, ?B/s]
- 19%|#9 | 7.99M/41.5M [00:00<00:00, 50.9MB/s]
- 35%|###4 | 14.3M/41.5M [00:00<00:00, 56.3MB/s]
- 48%|####7 | 19.8M/41.5M [00:00<00:00, 45.9MB/s]
- 59%|#####8 | 24.4M/41.5M [00:00<00:00, 24.8MB/s]
- 77%|#######7 | 32.0M/41.5M [00:01<00:00, 28.7MB/s]
- 85%|########5 | 35.3M/41.5M [00:01<00:00, 27.2MB/s]
- 92%|#########2| 38.3M/41.5M [00:01<00:00, 26.9MB/s]
-100%|##########| 41.5M/41.5M [00:01<00:00, 31.1MB/s]
+ 19%|#9 | 7.99M/41.5M [00:00<00:00, 53.4MB/s]
+ 39%|###8 | 16.0M/41.5M [00:00<00:00, 43.5MB/s]
+ 58%|#####7 | 24.0M/41.5M [00:00<00:00, 45.0MB/s]
+ 68%|######8 | 28.4M/41.5M [00:00<00:00, 44.1MB/s]
+ 78%|#######8 | 32.6M/41.5M [00:00<00:00, 43.4MB/s]
+ 92%|#########2| 38.3M/41.5M [00:00<00:00, 44.5MB/s]
+100%|##########| 41.5M/41.5M [00:00<00:00, 44.7MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index edb4a2234..d354e00c7 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -409,10 +409,9 @@ 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]
- 13%|#2 | 5.80M/44.7M [00:00<00:00, 60.8MB/s]
- 26%|##5 | 11.6M/44.7M [00:00<00:00, 58.5MB/s]
- 77%|#######6 | 34.2M/44.7M [00:00<00:00, 138MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 135MB/s]
+ 37%|###6 | 16.4M/44.7M [00:00<00:00, 172MB/s]
+ 86%|########5 | 38.3M/44.7M [00:00<00:00, 206MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 207MB/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 5fc4d6085..adb0b853d 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -631,7 +631,7 @@ 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 4.623 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 6.971 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-tensorflow-py">
<div class="sphx-glr-download sphx-glr-download-python 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 04dc2eb82..e1041e601 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -322,7 +322,7 @@
<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>05:04.184</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:11.910</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -331,43 +331,43 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><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></td>
-<td><p>01:04.623</p></td>
+<td><p>01:06.971</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>01:01.933</p></td>
+<td><p>01:03.724</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:39.601</p></td>
+<td><p>00:41.165</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_oneflow.html#sphx-glr-how-to-compile-models-from-oneflow-py"><span class="std std-ref">Compile OneFlow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_oneflow.py</span></code>)</p></td>
-<td><p>00:28.488</p></td>
+<td><p>00:27.598</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:25.761</p></td>
+<td><p>00:25.967</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:25.337</p></td>
+<td><p>00:25.928</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:21.884</p></td>
+<td><p>00:22.799</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:19.605</p></td>
+<td><p>00:20.387</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:14.552</p></td>
+<td><p>00:14.788</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:02.401</p></td>
+<td><p>00:02.583</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
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 2f543a6d9..cf8ea4a4e 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -648,7 +648,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.5391 15.5075 15.7231 15.4456 0.0831
+ 16.1154 15.9764 16.7093 15.8759 0.3033
</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 b7f59f993..695563312 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -431,14 +431,14 @@ 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]
- 10%|# | 17.7M/170M [00:00<00:00, 186MB/s]
- 24%|##4 | 41.0M/170M [00:00<00:00, 220MB/s]
- 38%|###8 | 64.8M/170M [00:00<00:00, 233MB/s]
- 52%|#####2 | 89.1M/170M [00:00<00:00, 242MB/s]
- 67%|######7 | 114M/170M [00:00<00:00, 250MB/s]
- 81%|########1 | 138M/170M [00:00<00:00, 245MB/s]
- 97%|#########6| 165M/170M [00:00<00:00, 255MB/s]
-100%|##########| 170M/170M [00:00<00:00, 245MB/s]
+ 9%|9 | 15.4M/170M [00:00<00:01, 161MB/s]
+ 21%|##1 | 36.5M/170M [00:00<00:00, 196MB/s]
+ 33%|###3 | 56.8M/170M [00:00<00:00, 204MB/s]
+ 47%|####6 | 79.2M/170M [00:00<00:00, 216MB/s]
+ 61%|###### | 103M/170M [00:00<00:00, 228MB/s]
+ 75%|#######4 | 127M/170M [00:00<00:00, 235MB/s]
+ 88%|########7 | 149M/170M [00:00<00:00, 236MB/s]
+100%|##########| 170M/170M [00:00<00:00, 226MB/s]
/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:3878: 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).
for i in range(dim)
/usr/local/lib/python3.7/dist-packages/torchvision/models/detection/anchor_utils.py:127: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -533,7 +533,7 @@ 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 51.868 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 0.852 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-object-detection-pytorch-py">
<div class="sphx-glr-download sphx-glr-download-python 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 062b4213e..b06a1be03 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -475,10 +475,7 @@ 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]
- 30%|##9 | 4.01M/13.6M [00:00<00:00, 41.8MB/s]
- 59%|#####9 | 8.00M/13.6M [00:00<00:00, 32.1MB/s]
- 90%|######### | 12.2M/13.6M [00:00<00:00, 36.9MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 37.9MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 158MB/s]
</pre></div>
</div>
</div>
@@ -567,7 +564,7 @@ output values are identical out of 1000 outputs from mobilenet v2.</p>
</div>
<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)
- 90.2425 90.0816 95.8397 89.8729 0.6945
+ 90.4531 90.2741 95.8757 90.1546 0.8351
</pre></div>
</div>
<div class="admonition note">
@@ -606,7 +603,7 @@ 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 6.952 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 10.607 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-py">
<div class="sphx-glr-download sphx-glr-download-python 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 f6416cbe2..8f2bed2c8 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -568,7 +568,7 @@ TFLite Top-5 labels: [387 102 386 341 349]
</div>
<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)
- 118.4942 118.5807 119.4968 116.9254 0.4745
+ 120.9951 120.9853 121.9960 120.0151 0.4267
</pre></div>
</div>
<div class="admonition note">
@@ -596,7 +596,7 @@ 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 57.893 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 58.945 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-tflite-py">
<div class="sphx-glr-download sphx-glr-download-python 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 3093b0521..744783ee7 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -504,7 +504,7 @@ 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 35.038 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 36.686 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-quantized-py">
<div class="sphx-glr-download sphx-glr-download-python 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 68c176366..99c7dace4 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -436,23 +436,24 @@ 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%|4 | 6416/132723 [00:00<00:01, 64148.44KB/s]
- 11%|#1 | 15072/132723 [00:00<00:01, 77327.85KB/s]
- 17%|#7 | 22805/132723 [00:00<00:01, 67826.06KB/s]
- 24%|##3 | 31402/132723 [00:00<00:01, 74426.31KB/s]
- 29%|##9 | 38974/132723 [00:00<00:01, 55477.01KB/s]
- 36%|###5 | 47595/132723 [00:00<00:01, 63644.67KB/s]
- 42%|####2 | 56247/132723 [00:00<00:01, 69939.60KB/s]
- 49%|####8 | 64980/132723 [00:00<00:00, 74842.81KB/s]
- 55%|#####4 | 72908/132723 [00:01<00:01, 58960.98KB/s]
- 61%|######1 | 81556/132723 [00:01<00:00, 65598.66KB/s]
- 67%|######6 | 88844/132723 [00:01<00:00, 62712.22KB/s]
- 74%|#######3 | 97556/132723 [00:01<00:00, 68909.03KB/s]
- 79%|#######9 | 104923/132723 [00:01<00:00, 57903.20KB/s]
- 86%|########5 | 113589/132723 [00:01<00:00, 64768.32KB/s]
- 91%|######### | 120658/132723 [00:01<00:00, 55674.31KB/s]
- 97%|#########7| 129350/132723 [00:02<00:00, 62981.16KB/s]
-100%|##########| 132723/132723 [00:02<00:00, 64465.36KB/s]
+ 4%|4 | 5714/132723 [00:00<00:02, 57136.38KB/s]
+ 10%|9 | 13185/132723 [00:00<00:01, 67470.45KB/s]
+ 16%|#5 | 20825/132723 [00:00<00:01, 71543.85KB/s]
+ 21%|##1 | 28474/132723 [00:00<00:01, 73495.14KB/s]
+ 27%|##7 | 36085/132723 [00:00<00:01, 74434.34KB/s]
+ 33%|###2 | 43675/132723 [00:00<00:01, 74930.05KB/s]
+ 39%|###8 | 51196/132723 [00:00<00:01, 75020.07KB/s]
+ 44%|####4 | 58809/132723 [00:00<00:00, 75371.70KB/s]
+ 50%|##### | 66475/132723 [00:00<00:00, 75773.26KB/s]
+ 56%|#####5 | 74161/132723 [00:01<00:00, 76106.02KB/s]
+ 62%|######1 | 81873/132723 [00:01<00:00, 76412.95KB/s]
+ 67%|######7 | 89515/132723 [00:01<00:00, 70234.51KB/s]
+ 73%|#######3 | 97065/132723 [00:01<00:00, 71734.88KB/s]
+ 79%|#######8 | 104609/132723 [00:01<00:00, 72806.05KB/s]
+ 85%|########4 | 112289/132723 [00:01<00:00, 73971.53KB/s]
+ 90%|######### | 119731/132723 [00:01<00:00, 65394.26KB/s]
+ 96%|#########6| 127456/132723 [00:01<00:00, 68605.27KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 71998.12KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -495,7 +496,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 29.778 seconds)</p>
+<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 38.979 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-ssd-gluoncv-py">
<div class="sphx-glr-download sphx-glr-download-python 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 d7b1766da..f1ea51e67 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -322,7 +322,7 @@
<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>10:52.592</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>11:19.030</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -331,31 +331,31 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><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></td>
-<td><p>02:51.868</p></td>
+<td><p>03:00.852</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>02:29.778</p></td>
+<td><p>02:38.979</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>01:57.893</p></td>
+<td><p>01:58.945</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>01:35.038</p></td>
+<td><p>01:36.686</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>01:06.952</p></td>
+<td><p>01:10.607</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:28.604</p></td>
+<td><p>00:29.824</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:22.454</p></td>
+<td><p>00:23.130</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
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 531a81ec9..6c06db4b1 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -607,7 +607,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
<span class="n">module</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">params</span></a> <span class="o">=</span> <span class="n">get_mobilenet</span><span class="p">()</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip8a4ec15b-dd30-4708-9e57-4287dc08f05d 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.zip16beb6ff-04ac-4110-a3ad-46f0a7b37d55 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>
diff --git a/docs/how_to/extend_tvm/sg_execution_times.html b/docs/how_to/extend_tvm/sg_execution_times.html
index 6200bb193..f352aa398 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -322,7 +322,7 @@
<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>00:39.491</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:41.583</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -331,19 +331,19 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:36.443</p></td>
+<td><p>00:38.303</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:02.141</p></td>
+<td><p>00:02.309</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:00.900</p></td>
+<td><p>00:00.961</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:00.007</p></td>
+<td><p>00:00.009</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index 7fea98dd6..8ff86ef47 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -507,10 +507,10 @@ profile the execution time of each passes.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 6404us [6404us] (45.59%; 45.59%)
-FoldScaleAxis: 7644us [5us] (54.41%; 54.41%)
- FoldConstant: 7639us [1587us] (54.38%; 99.94%)
- InferType: 6053us [6053us] (43.08%; 79.23%)
+InferType: 6829us [6829us] (45.29%; 45.29%)
+FoldScaleAxis: 8250us [7us] (54.71%; 54.71%)
+ FoldConstant: 8243us [1615us] (54.67%; 99.92%)
+ InferType: 6628us [6628us] (43.96%; 80.41%)
</pre></div>
</div>
</div>
@@ -532,10 +532,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 6125us [6125us] (44.64%; 44.64%)
-FoldScaleAxis: 7596us [4us] (55.36%; 55.36%)
- FoldConstant: 7592us [1567us] (55.33%; 99.94%)
- InferType: 6025us [6025us] (43.91%; 79.36%)
+InferType: 6643us [6643us] (44.83%; 44.83%)
+FoldScaleAxis: 8176us [7us] (55.17%; 55.17%)
+ FoldConstant: 8169us [1704us] (55.13%; 99.92%)
+ InferType: 6466us [6466us] (43.63%; 79.15%)
</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 f90cb6cd6..f77e1846c 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -559,7 +559,7 @@ latency of convolution.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Convolution: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">*</span> <span cl [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 38.155233 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 54.298894 ms
</pre></div>
</div>
<div class="sphx-glr-footer 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 aefe440ab..6ac4ea03b 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -901,7 +901,7 @@ be able to run on our build server</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"conv2d with tensor core: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">* [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 7.812188 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 7.178832 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 b460f6942..83d643aa9 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -456,8 +456,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
<span class="nb">print</span><span class="p">(</span><span class="s2">"Baseline: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018373
-Baseline: 3.191560
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.020718
+Baseline: 3.453568
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -517,7 +517,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt1: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.289777
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.319622
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -584,7 +584,7 @@ vastly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt2: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.331577
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.345024
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -645,7 +645,7 @@ the access pattern for A matrix is more cache friendly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt3: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.117284
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.122990
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -728,7 +728,7 @@ flattening.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt4: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.109667
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.111755
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -814,7 +814,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt5: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.110944
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111711
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -904,7 +904,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt6: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">opt6_time</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.144450
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.145288
</pre></div>
</div>
<p>Here is the generated IR after parallelization.</p>
diff --git a/docs/how_to/optimize_operators/sg_execution_times.html b/docs/how_to/optimize_operators/sg_execution_times.html
index a678c39c2..b7b2f32c3 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -322,7 +322,7 @@
<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>00:33.734</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.361</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -331,15 +331,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:31.404</p></td>
+<td><p>00:33.031</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:01.294</p></td>
+<td><p>00:01.279</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:01.036</p></td>
+<td><p>00:01.050</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
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 b3659e6f9..c09b0ae2e 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -322,7 +322,7 @@
<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>05:58.163</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>06:09.006</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -331,27 +331,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><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></td>
-<td><p>03:13.851</p></td>
+<td><p>03:21.685</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>01:21.130</p></td>
+<td><p>01:23.406</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:45.044</p></td>
+<td><p>00:46.366</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:21.043</p></td>
+<td><p>00:19.544</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:08.608</p></td>
+<td><p>00:09.145</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:08.487</p></td>
+<td><p>00:08.860</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
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 e5352fd11..d1d75c25b 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
@@ -486,370 +486,1119 @@ cooperative fetching, unrolling and operator fusion.</p>
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 64;
- allocate(conv2d_nchw: Pointer(local float32), float32, [4]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [504]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [192]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [4], [], 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, [28]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [324]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [1152]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [196], [], scope="local", align=32)[0] = 0f32
+ conv2d_nchw_1[14] = 0f32
conv2d_nchw_1[1] = 0f32
+ conv2d_nchw_1[15] = 0f32
conv2d_nchw_1[2] = 0f32
+ conv2d_nchw_1[16] = 0f32
conv2d_nchw_1[3] = 0f32
- for (rc.outer.outer: int32, 0, 64) {
- let cse_var_1: int32 = (rc.outer.outer*392)
+ conv2d_nchw_1[17] = 0f32
+ conv2d_nchw_1[4] = 0f32
+ conv2d_nchw_1[18] = 0f32
+ conv2d_nchw_1[5] = 0f32
+ conv2d_nchw_1[19] = 0f32
+ conv2d_nchw_1[6] = 0f32
+ conv2d_nchw_1[20] = 0f32
+ conv2d_nchw_1[7] = 0f32
+ conv2d_nchw_1[21] = 0f32
+ conv2d_nchw_1[8] = 0f32
+ conv2d_nchw_1[22] = 0f32
+ conv2d_nchw_1[9] = 0f32
+ conv2d_nchw_1[23] = 0f32
+ conv2d_nchw_1[10] = 0f32
+ conv2d_nchw_1[24] = 0f32
+ conv2d_nchw_1[11] = 0f32
+ conv2d_nchw_1[25] = 0f32
+ conv2d_nchw_1[12] = 0f32
+ conv2d_nchw_1[26] = 0f32
+ conv2d_nchw_1[13] = 0f32
+ conv2d_nchw_1[27] = 0f32
+ for (rc.outer.outer: int32, 0, 128) {
+ let cse_var_2: int32 = (rc.outer.outer*196)
+ let cse_var_1: int32 = (rc.outer.outer*36)
{
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [504], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (1 <= floormod(threadIdx.x_1, 7))), data[(((cse_var_1 + (floordiv(threadIdx.x_1, 63)*49)) + floormod(threadIdx.x_1, 63)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 1), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 294), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_1 < 14), dtype=bool) {
- pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else(((threadIdx.x_1 < 7) && (1 <= floormod(threadIdx.x_1, 7))), data[(((cse_var_1 + (floordiv((threadIdx.x_1 + 490), 63)*49)) + threadIdx.x_1) + 41)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [324], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else((((9 <= threadIdx.x_1) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[(((cse_var_2 + (floordiv(threadIdx.x_1, 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 56)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 56), 81)) && (floormod((threadIdx.x_1 + 56), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 56), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 56), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 31), 81)) && (floormod((threadIdx.x_1 + 31), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 112), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 31), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 168)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 6), 81)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 168), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 6), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 62), 81)) && (floormod((threadIdx.x_1 + 62), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 224), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 62), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ if @tir.likely((threadIdx.x_1 < 44), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 280)] = @tir.if_then_else((((threadIdx.x_1 < 35) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 280), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 37), 81), 9)*7)) + floormod((threadIdx.x_1 + 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, [192], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 24)*3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_2 < 94), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 98), 24)*4608)) + (rc.outer.outer*72)) + (floormod((threadIdx.x_2 + 2), 24)*3))]
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1: Buffer(kernel.shared, float32, [1152], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 36)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 56), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 20), 36), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 112), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 4), 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 168), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 24), 36), 9)*9)) + floormod((threadIdx.x_2 + 6), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 224), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 280)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 280), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 28), 36), 9)*9)) + floormod((threadIdx.x_2 + 1), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 336), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 12), 36), 9)*9)) + floormod((threadIdx.x_2 + 3), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 392), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 36), 9)*9)) + floormod((threadIdx.x_2 + 5), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 448), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 36), 9)*9)) + floormod((threadIdx.x_2 + 7), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 504)] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 36)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 36)) + 64512)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 560), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 20), 36), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 616)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 616), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 4), 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 672), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 24), 36), 9)*9)) + floormod((threadIdx.x_2 + 6), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 728)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 728), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 36))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 784), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 28), 36), 9)*9)) + floormod((threadIdx.x_2 + 1), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 840)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 840), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 12), 36), 9)*9)) + floormod((threadIdx.x_2 + 3), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 896), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 36), 9)*9)) + floormod((threadIdx.x_2 + 5), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 952)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 952), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 36), 9)*9)) + floormod((threadIdx.x_2 + 7), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1008)] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 36)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 36)) + 129024)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1064)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1064), 36)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 20), 36), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel[((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1120), 36)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 4), 36))]
}
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 95)]))
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)), data[(((cse_var_1 + (floordiv(threadIdx.x_1, 63)*49)) + floormod(threadIdx.x_1, 63)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 1), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 294), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_1 < 14), dtype=bool) {
- pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else((threadIdx.x_1 < 7), data[(((cse_var_1 + (floordiv((threadIdx.x_1 + 490), 63)*49)) + threadIdx.x_1) + 42)], 0f32, dtype=float32)
- }
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 24)*3)) + 1)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_2 < 94), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 98), 24)*4608)) + (rc.outer.outer*72)) + (floormod((threadIdx.x_2 + 2), 24)*3)) + 1)]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 95)]))
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (floormod(threadIdx.x_1, 7) < 6)), data[(((cse_var_1 + (floordiv(threadIdx.x_1, 63)*49)) + floormod(threadIdx.x_1, 63)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 1), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 294), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_1 < 14), dtype=bool) {
- pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else(((threadIdx.x_1 < 7) && (floormod(threadIdx.x_1, 7) < 6)), data[(((cse_var_1 + (floordiv((threadIdx.x_1 + 490), 63)*49)) + threadIdx.x_1) + 43)], 0f32, dtype=float32)
- }
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 24)*3)) + 2)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
- if @tir.likely((threadIdx.x_2 < 94), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 98), 24)*4608)) + (rc.outer.outer*72)) + (floormod((threadIdx.x_2 + 2), 24)*3)) + 2)]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 95)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*72)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 576)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 1)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 577)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 2)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 578)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 36)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 612)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 37)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 613)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 38)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 614)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 3)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 579)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 4)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 580)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 5)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 581)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 10)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 11)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 39)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 615)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 40)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 616)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 41)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 617)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 6)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 582)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 7)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 583)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 8)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 584)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 18)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 19)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 20)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 36)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 37)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 38)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 45)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 46)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 47)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 54)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 55)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 42)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 72)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 618)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 43)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 73)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 619)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 44)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 74)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 620)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 9)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 585)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 10)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 586)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 11)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 587)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 81)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 82)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 83)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 45)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 621)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 46)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 622)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 47)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 623)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 12)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 588)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 13)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 589)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 14)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 590)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 90)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 92)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 48)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 624)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 49)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 625)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 50)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 626)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 15)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 591)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 16)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 592)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 17)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 593)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 99)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 100)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 101)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 108)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 109)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 110)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 117)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 118)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 135)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 136)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 137)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 144)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 145)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 146)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 51)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 153)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 627)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 52)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 628)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 53)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 155)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 629)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 18)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 594)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 19)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 595)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 20)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 596)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 162)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 163)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 164)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 54)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 630)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 55)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 631)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 56)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 632)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 21)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 597)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 22)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 598)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 23)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 599)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 171)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 172)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 173)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 57)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 633)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 58)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 634)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 59)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 635)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 24)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 600)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 25)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 601)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 26)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 602)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 180)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 181)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 198)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 199)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 200)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 207)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 208)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 209)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 216)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 218)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 225)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 226)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 227)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 60)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 234)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 636)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 61)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 235)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 637)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 62)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 236)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 638)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 27)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 603)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 28)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 604)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 29)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 605)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 243)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 244)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 63)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 639)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 64)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 640)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 65)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 641)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 30)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 606)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 31)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 607)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 32)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 608)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 66)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 642)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 67)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 643)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 68)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 644)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 33)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 609)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 34)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 610)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 35)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 611)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 261)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 262)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 263)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 270)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 271)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 272)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 279)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 281)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 288)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 289)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 290)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 297)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 298)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 299)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 306)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 307)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 69)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 645)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 70)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 646)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 71)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + 647)]))
}
}
- for (i1.inner: int32, 0, 4) {
- compute[((((blockIdx.x*392) + (floordiv(threadIdx.x, 49)*196)) + (i1.inner*49)) + floormod(threadIdx.x, 49))] = max((conv2d_nchw_1[i1.inner] + bias[(((blockIdx.x*8) + (floordiv(threadIdx.x, 49)*4)) + i1.inner)]), 0f32)
+ for (i1.inner: int32, 0, 2) {
+ for (i2.inner: int32, 0, 7) {
+ let cse_var_3: int32 = ((i1.inner*7) + i2.inner)
+ {
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[cse_var_3] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner)]), 0f32)
+ compute[((((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7)) + 784)] = max((conv2d_nchw_1[(cse_var_3 + 14)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner) + 16)]), 0f32)
+ }
+ }
}
}
}
@@ -886,7 +1635,7 @@ cooperative fetching, unrolling and operator fusion.</p>
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.247 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.377 ms
</pre></div>
</div>
</div>
@@ -916,32 +1665,32 @@ conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o
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=4)
-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_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=8)
+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=2)
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=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_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_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=1)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=1)
+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_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=4)
-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=1)
-compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=8)
+compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
+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_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)
@@ -964,12 +1713,12 @@ 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=98)
+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=56)
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=1)
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=98)
+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=56)
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)
@@ -989,348 +1738,1086 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-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[4];
- __shared__ float pad_temp_shared[504];
- __shared__ float kernel_shared[192];
+extern "C" __global__ void __launch_bounds__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[28];
+ __shared__ float pad_temp_shared[324];
+ __shared__ float kernel_shared[1152];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[16] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
+ conv2d_nchw[17] = 0.000000e+00f;
+ conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[18] = 0.000000e+00f;
+ conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[19] = 0.000000e+00f;
+ conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[20] = 0.000000e+00f;
+ conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[21] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[22] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[23] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[24] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[25] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[26] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
+ conv2d_nchw[27] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 128; ++rc_outer_outer) {
__syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = ((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && (1 <= (((int)threadIdx.x) % 7))) ? data[((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 63) * 49)) + (((int)threadIdx.x) % 63)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 98)] = ((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 196)] = ((((1 <= (((((int)threadIdx.x) / 7) + 1) % 9)) && ((((((int)threadIdx.x) / 7) + 1) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 1) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 294)] = ((((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 294) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = ((((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 14) {
- pad_temp_shared[(((int)threadIdx.x) + 490)] = (((((int)threadIdx.x) < 7) && (1 <= (((int)threadIdx.x) % 7))) ? data[((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 490) / 63) * 49)) + ((int)threadIdx.x)) + 41)] : 0.000000e+00f);
+ pad_temp_shared[((int)threadIdx.x)] = ((((9 <= ((int)threadIdx.x)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[((((rc_outer_outer * 196) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 56)] = (((((9 <= ((((int)threadIdx.x) + 56) % 81)) && (((((int)threadIdx.x) + 56) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 56) / 81) * 49)) + ((((((int)threadIdx.x) + 56) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((9 <= ((((int)threadIdx.x) + 31) % 81)) && (((((int)threadIdx.x) + 31) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 112) / 81) * 49)) + ((((((int)threadIdx.x) + 31) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 168)] = ((((3 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 168) / 81) * 49)) + (((((int)threadIdx.x) + 6) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((9 <= ((((int)threadIdx.x) + 62) % 81)) && (((((int)threadIdx.x) + 62) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 224) / 81) * 49)) + ((((((int)threadIdx.x) + 62) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 44) {
+ pad_temp_shared[(((int)threadIdx.x) + 280)] = ((((((int)threadIdx.x) < 35) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 280) / 81) * 49)) + (((((int)threadIdx.x) + 37) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
}
- kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3))];
- if (((int)threadIdx.x) < 94) {
- kernel_shared[(((int)threadIdx.x) + 98)] = kernel[((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 98) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 2) % 24) * 3))];
+ kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 36) * 4608)) + (rc_outer_outer * 36)) + (((int)threadIdx.x) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 56)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 20) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 2) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) + 4) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 24) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 6) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) + 8) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 280)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 28) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 1) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 12) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 3) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 32) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 5) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 16) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 7) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 504)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 36) * 4608)) + (rc_outer_outer * 36)) + (((int)threadIdx.x) % 36)) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 20) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 2) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 616)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 616) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) + 4) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 672) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 24) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 6) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 728)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 728) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) + 8) % 36))];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 28) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 1) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 840)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 840) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 12) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 3) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 896) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 32) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 5) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 952) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 16) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 7) % 9))];
+ kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 36) * 4608)) + (rc_outer_outer * 36)) + (((int)threadIdx.x) % 36)) + 129024)];
+ kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1064) / 36) * 4608)) + (rc_outer_outer * 36)) + ((((((int)threadIdx.x) + 20) % 36) / 9) * 9)) + ((((int)threadIdx.x) + 2) % 9))];
+ if (((int)threadIdx.x) < 32) {
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1120) / 36) * 4608)) + (rc_outer_outer * 36)) + (((int)threadIdx.x) + 4))];
}
__syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 95)]));
- __syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = (((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) ? data[((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 63) * 49)) + (((int)threadIdx.x) % 63)) - 7)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 98)] = (((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 196)] = (((1 <= (((((int)threadIdx.x) / 7) + 1) % 9)) && ((((((int)threadIdx.x) / 7) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 1) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 294)] = (((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 294) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = (((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 14) {
- pad_temp_shared[(((int)threadIdx.x) + 490)] = ((((int)threadIdx.x) < 7) ? data[((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 490) / 63) * 49)) + ((int)threadIdx.x)) + 42)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + 1)];
- if (((int)threadIdx.x) < 94) {
- kernel_shared[(((int)threadIdx.x) + 98)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 98) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 2) % 24) * 3)) + 1)];
- }
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 95)]));
- __syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = ((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && ((((int)threadIdx.x) % 7) < 6)) ? data[((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 63) * 49)) + (((int)threadIdx.x) % 63)) - 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 98)] = ((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 196)] = ((((1 <= (((((int)threadIdx.x) / 7) + 1) % 9)) && ((((((int)threadIdx.x) / 7) + 1) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 1) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 294)] = ((((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 294) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = ((((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 14) {
- pad_temp_shared[(((int)threadIdx.x) + 490)] = (((((int)threadIdx.x) < 7) && ((((int)threadIdx.x) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 490) / 63) * 49)) + ((int)threadIdx.x)) + 43)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + 2)];
- if (((int)threadIdx.x) < 94) {
- kernel_shared[(((int)threadIdx.x) + 98)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 98) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 2) % 24) * 3)) + 2)];
- }
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 95)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[((((int)threadIdx.x) / 7) * 72)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 576)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 1)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 577)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 2)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 578)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 36)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 612)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 37)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 613)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 38)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 614)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 3)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 579)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 4)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 580)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 5)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 581)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 10)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 11)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 39)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 615)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 40)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 616)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 41)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 617)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 6)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 582)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 7)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 583)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 8)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 584)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 18)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 19)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 20)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 36)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 37)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 38)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 45)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 46)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 47)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 54)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 55)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 42)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 72)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 618)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 43)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 73)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 619)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 44)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 74)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 620)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 9)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 585)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 10)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 586)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 11)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 587)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 81)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 82)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 83)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 45)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 621)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 46)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 622)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 47)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 623)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 12)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 588)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 13)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 589)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 14)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 590)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 90)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 92)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 48)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 624)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 49)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 625)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 50)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 626)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 15)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 591)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 16)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 592)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 17)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 593)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 99)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 100)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 101)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 108)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 109)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 110)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 117)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 118)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 135)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 136)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 137)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 144)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 145)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 146)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 51)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 153)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 627)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 52)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 628)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 53)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 155)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 629)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 18)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 594)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 19)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 595)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 20)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 596)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 162)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 163)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 164)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 54)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 630)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 55)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 631)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 56)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 632)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 21)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 597)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 22)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 598)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 23)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 599)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 171)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 172)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 173)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 57)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 633)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 58)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 634)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 59)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 635)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 24)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 600)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 25)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 601)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 26)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 602)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 180)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 181)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 198)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 199)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 200)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 207)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 208)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 209)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 216)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 218)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 225)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 226)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 227)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 60)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 234)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 636)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 61)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 235)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 637)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 62)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 236)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 638)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 27)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 603)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 28)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 604)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 29)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 605)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 243)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 244)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 63)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 639)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 64)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 640)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 65)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 641)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 30)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 606)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 31)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 607)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 32)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 608)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 66)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 642)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 67)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 643)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 68)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 644)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 33)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 609)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 34)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 610)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 35)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 611)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 261)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 262)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 263)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 270)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 271)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 272)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 279)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 281)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 288)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 289)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 290)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 297)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 298)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 299)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 306)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 307)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 69)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 645)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 70)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 646)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 71)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + 647)]));
}
- for (int i1_inner = 0; i1_inner < 4; ++i1_inner) {
- compute[((((((int)blockIdx.x) * 392) + ((((int)threadIdx.x) / 49) * 196)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 8) + ((((int)threadIdx.x) / 49) * 4)) + i1_inner)]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+ for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ compute[((((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7)) + 784)] = max((conv2d_nchw[(((i1_inner * 7) + i2_inner) + 14)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner) + 16)]), 0.000000e+00f);
+ }
}
}
</pre></div>
@@ -1367,7 +2854,7 @@ 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> ( 3 minutes 13.851 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 21.685 seconds)</p>
<div class="sphx-glr-footer 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 sphx-glr-download-python 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 8e49ece8f..cd1da47dc 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -901,7 +901,7 @@ 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)
- 9.8203 9.7996 9.8876 9.7736 0.0488
+ 9.6666 9.6901 9.7160 9.5937 0.0526
</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 df95f4b8f..4989dd4c6 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -920,7 +920,7 @@ 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)
- 751.6063 751.6842 751.8405 751.2941 0.2297
+ 763.3527 763.4943 764.3378 762.2261 0.8679
</pre></div>
</div>
</div>
@@ -942,7 +942,7 @@ 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 21.130 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 23.406 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-network-x86-py">
<div class="sphx-glr-download sphx-glr-download-python 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 d8834215a..a592ba575 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -620,407 +620,103 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_9: placeholder_15: Buffer(placeholder_14, float32, [128, 512], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_7: placeholder_16: Buffer(placeholder_12, int32, [4916], []), placeholder_8: placeholder_17: Buffer(placeholder_13, int32, [33], []), placeholder_5: placeholder_18: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 256) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [256]), storage_scope = global {
- for (nb_j.inner: int32, 0, 2) {
- let cse_var_2: int32 = (nb_j.inner*16)
- let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
- {
- compute_5: Buffer(compute_4, float32, [256], [])[cse_var_2] = 0f32
- compute_5[(cse_var_2 + 1)] = 0f32
- compute_5[(cse_var_2 + 2)] = 0f32
- compute_5[(cse_var_2 + 3)] = 0f32
- compute_5[(cse_var_2 + 4)] = 0f32
- compute_5[(cse_var_2 + 5)] = 0f32
- compute_5[(cse_var_2 + 6)] = 0f32
- compute_5[(cse_var_2 + 7)] = 0f32
- compute_5[(cse_var_2 + 8)] = 0f32
- compute_5[(cse_var_2 + 9)] = 0f32
- compute_5[(cse_var_2 + 10)] = 0f32
- compute_5[(cse_var_2 + 11)] = 0f32
- compute_5[(cse_var_2 + 12)] = 0f32
- compute_5[(cse_var_2 + 13)] = 0f32
- compute_5[(cse_var_2 + 14)] = 0f32
- compute_5[(cse_var_2 + 15)] = 0f32
- compute_5[(cse_var_2 + 32)] = 0f32
- compute_5[(cse_var_2 + 33)] = 0f32
- compute_5[(cse_var_2 + 34)] = 0f32
- compute_5[(cse_var_2 + 35)] = 0f32
- compute_5[(cse_var_2 + 36)] = 0f32
- compute_5[(cse_var_2 + 37)] = 0f32
- compute_5[(cse_var_2 + 38)] = 0f32
- compute_5[(cse_var_2 + 39)] = 0f32
- compute_5[(cse_var_2 + 40)] = 0f32
- compute_5[(cse_var_2 + 41)] = 0f32
- compute_5[(cse_var_2 + 42)] = 0f32
- compute_5[(cse_var_2 + 43)] = 0f32
- compute_5[(cse_var_2 + 44)] = 0f32
- compute_5[(cse_var_2 + 45)] = 0f32
- compute_5[(cse_var_2 + 46)] = 0f32
- compute_5[(cse_var_2 + 47)] = 0f32
- compute_5[(cse_var_2 + 64)] = 0f32
- compute_5[(cse_var_2 + 65)] = 0f32
- compute_5[(cse_var_2 + 66)] = 0f32
- compute_5[(cse_var_2 + 67)] = 0f32
- compute_5[(cse_var_2 + 68)] = 0f32
- compute_5[(cse_var_2 + 69)] = 0f32
- compute_5[(cse_var_2 + 70)] = 0f32
- compute_5[(cse_var_2 + 71)] = 0f32
- compute_5[(cse_var_2 + 72)] = 0f32
- compute_5[(cse_var_2 + 73)] = 0f32
- compute_5[(cse_var_2 + 74)] = 0f32
- compute_5[(cse_var_2 + 75)] = 0f32
- compute_5[(cse_var_2 + 76)] = 0f32
- compute_5[(cse_var_2 + 77)] = 0f32
- compute_5[(cse_var_2 + 78)] = 0f32
- compute_5[(cse_var_2 + 79)] = 0f32
- compute_5[(cse_var_2 + 96)] = 0f32
- compute_5[(cse_var_2 + 97)] = 0f32
- compute_5[(cse_var_2 + 98)] = 0f32
- compute_5[(cse_var_2 + 99)] = 0f32
- compute_5[(cse_var_2 + 100)] = 0f32
- compute_5[(cse_var_2 + 101)] = 0f32
- compute_5[(cse_var_2 + 102)] = 0f32
- compute_5[(cse_var_2 + 103)] = 0f32
- compute_5[(cse_var_2 + 104)] = 0f32
- compute_5[(cse_var_2 + 105)] = 0f32
- compute_5[(cse_var_2 + 106)] = 0f32
- compute_5[(cse_var_2 + 107)] = 0f32
- compute_5[(cse_var_2 + 108)] = 0f32
- compute_5[(cse_var_2 + 109)] = 0f32
- compute_5[(cse_var_2 + 110)] = 0f32
- compute_5[(cse_var_2 + 111)] = 0f32
- compute_5[(cse_var_2 + 128)] = 0f32
- compute_5[(cse_var_2 + 129)] = 0f32
- compute_5[(cse_var_2 + 130)] = 0f32
- compute_5[(cse_var_2 + 131)] = 0f32
- compute_5[(cse_var_2 + 132)] = 0f32
- compute_5[(cse_var_2 + 133)] = 0f32
- compute_5[(cse_var_2 + 134)] = 0f32
- compute_5[(cse_var_2 + 135)] = 0f32
- compute_5[(cse_var_2 + 136)] = 0f32
- compute_5[(cse_var_2 + 137)] = 0f32
- compute_5[(cse_var_2 + 138)] = 0f32
- compute_5[(cse_var_2 + 139)] = 0f32
- compute_5[(cse_var_2 + 140)] = 0f32
- compute_5[(cse_var_2 + 141)] = 0f32
- compute_5[(cse_var_2 + 142)] = 0f32
- compute_5[(cse_var_2 + 143)] = 0f32
- compute_5[(cse_var_2 + 160)] = 0f32
- compute_5[(cse_var_2 + 161)] = 0f32
- compute_5[(cse_var_2 + 162)] = 0f32
- compute_5[(cse_var_2 + 163)] = 0f32
- compute_5[(cse_var_2 + 164)] = 0f32
- compute_5[(cse_var_2 + 165)] = 0f32
- compute_5[(cse_var_2 + 166)] = 0f32
- compute_5[(cse_var_2 + 167)] = 0f32
- compute_5[(cse_var_2 + 168)] = 0f32
- compute_5[(cse_var_2 + 169)] = 0f32
- compute_5[(cse_var_2 + 170)] = 0f32
- compute_5[(cse_var_2 + 171)] = 0f32
- compute_5[(cse_var_2 + 172)] = 0f32
- compute_5[(cse_var_2 + 173)] = 0f32
- compute_5[(cse_var_2 + 174)] = 0f32
- compute_5[(cse_var_2 + 175)] = 0f32
- compute_5[(cse_var_2 + 192)] = 0f32
- compute_5[(cse_var_2 + 193)] = 0f32
- compute_5[(cse_var_2 + 194)] = 0f32
- compute_5[(cse_var_2 + 195)] = 0f32
- compute_5[(cse_var_2 + 196)] = 0f32
- compute_5[(cse_var_2 + 197)] = 0f32
- compute_5[(cse_var_2 + 198)] = 0f32
- compute_5[(cse_var_2 + 199)] = 0f32
- compute_5[(cse_var_2 + 200)] = 0f32
- compute_5[(cse_var_2 + 201)] = 0f32
- compute_5[(cse_var_2 + 202)] = 0f32
- compute_5[(cse_var_2 + 203)] = 0f32
- compute_5[(cse_var_2 + 204)] = 0f32
- compute_5[(cse_var_2 + 205)] = 0f32
- compute_5[(cse_var_2 + 206)] = 0f32
- compute_5[(cse_var_2 + 207)] = 0f32
- compute_5[(cse_var_2 + 224)] = 0f32
- compute_5[(cse_var_2 + 225)] = 0f32
- compute_5[(cse_var_2 + 226)] = 0f32
- compute_5[(cse_var_2 + 227)] = 0f32
- compute_5[(cse_var_2 + 228)] = 0f32
- compute_5[(cse_var_2 + 229)] = 0f32
- compute_5[(cse_var_2 + 230)] = 0f32
- compute_5[(cse_var_2 + 231)] = 0f32
- compute_5[(cse_var_2 + 232)] = 0f32
- compute_5[(cse_var_2 + 233)] = 0f32
- compute_5[(cse_var_2 + 234)] = 0f32
- compute_5[(cse_var_2 + 235)] = 0f32
- compute_5[(cse_var_2 + 236)] = 0f32
- compute_5[(cse_var_2 + 237)] = 0f32
- compute_5[(cse_var_2 + 238)] = 0f32
- compute_5[(cse_var_2 + 239)] = 0f32
- for (elem_idx: int32, 0, (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- let cse_var_131: int32 = (elem_idx*16)
- let cse_var_130: int32 = (cse_var_2 + 99)
- let cse_var_129: int32 = (cse_var_2 + 98)
- let cse_var_128: int32 = (cse_var_2 + 97)
- let cse_var_127: int32 = (cse_var_2 + 96)
- let cse_var_126: int32 = (cse_var_2 + 9)
- let cse_var_125: int32 = (cse_var_2 + 8)
- let cse_var_124: int32 = (cse_var_2 + 79)
- let cse_var_123: int32 = (cse_var_2 + 78)
- let cse_var_122: int32 = (cse_var_2 + 77)
- let cse_var_121: int32 = (cse_var_2 + 76)
- let cse_var_120: int32 = (cse_var_2 + 75)
- let cse_var_119: int32 = (cse_var_2 + 74)
- let cse_var_118: int32 = (cse_var_2 + 73)
- let cse_var_117: int32 = (cse_var_2 + 72)
- let cse_var_116: int32 = (cse_var_2 + 71)
- let cse_var_115: int32 = (cse_var_2 + 70)
- let cse_var_114: int32 = (cse_var_2 + 7)
- let cse_var_113: int32 = (cse_var_2 + 69)
- let cse_var_112: int32 = (cse_var_2 + 68)
- let cse_var_111: int32 = (cse_var_2 + 67)
- let cse_var_110: int32 = (cse_var_2 + 66)
- let cse_var_109: int32 = (cse_var_2 + 65)
- let cse_var_108: int32 = (cse_var_2 + 64)
- let cse_var_107: int32 = (cse_var_2 + 6)
- let cse_var_106: int32 = (cse_var_2 + 5)
- let cse_var_105: int32 = (cse_var_2 + 47)
- let cse_var_104: int32 = (cse_var_2 + 46)
- let cse_var_103: int32 = (cse_var_2 + 45)
- let cse_var_102: int32 = (cse_var_2 + 44)
- let cse_var_101: int32 = (cse_var_2 + 43)
- let cse_var_100: int32 = (cse_var_2 + 42)
- let cse_var_99: int32 = (cse_var_2 + 41)
- let cse_var_98: int32 = (cse_var_2 + 40)
- let cse_var_97: int32 = (cse_var_2 + 4)
- let cse_var_96: int32 = (cse_var_2 + 39)
- let cse_var_95: int32 = (cse_var_2 + 38)
- let cse_var_94: int32 = (cse_var_2 + 37)
- let cse_var_93: int32 = (cse_var_2 + 36)
- let cse_var_92: int32 = (cse_var_2 + 35)
- let cse_var_91: int32 = (cse_var_2 + 34)
- let cse_var_90: int32 = (cse_var_2 + 33)
- let cse_var_89: int32 = (cse_var_2 + 32)
- let cse_var_88: int32 = (cse_var_2 + 3)
- let cse_var_87: int32 = (cse_var_2 + 239)
- let cse_var_86: int32 = (cse_var_2 + 238)
- let cse_var_85: int32 = (cse_var_2 + 237)
- let cse_var_84: int32 = (cse_var_2 + 236)
- let cse_var_83: int32 = (cse_var_2 + 235)
- let cse_var_82: int32 = (cse_var_2 + 234)
- let cse_var_81: int32 = (cse_var_2 + 233)
- let cse_var_80: int32 = (cse_var_2 + 232)
- let cse_var_79: int32 = (cse_var_2 + 231)
- let cse_var_78: int32 = (cse_var_2 + 230)
- let cse_var_77: int32 = (cse_var_2 + 229)
- let cse_var_76: int32 = (cse_var_2 + 228)
- let cse_var_75: int32 = (cse_var_2 + 227)
- let cse_var_74: int32 = (cse_var_2 + 226)
- let cse_var_73: int32 = (cse_var_2 + 225)
- let cse_var_72: int32 = (cse_var_2 + 224)
- let cse_var_71: int32 = (cse_var_2 + 207)
- let cse_var_70: int32 = (cse_var_2 + 206)
- let cse_var_69: int32 = (cse_var_2 + 205)
- let cse_var_68: int32 = (cse_var_2 + 204)
- let cse_var_67: int32 = (cse_var_2 + 203)
- let cse_var_66: int32 = (cse_var_2 + 202)
- let cse_var_65: int32 = (cse_var_2 + 201)
- let cse_var_64: int32 = (cse_var_2 + 200)
- let cse_var_63: int32 = (cse_var_2 + 2)
- let cse_var_62: int32 = (cse_var_2 + 199)
- let cse_var_61: int32 = (cse_var_2 + 198)
- let cse_var_60: int32 = (cse_var_2 + 197)
- let cse_var_59: int32 = (cse_var_2 + 196)
- let cse_var_58: int32 = (cse_var_2 + 195)
- let cse_var_57: int32 = (cse_var_2 + 194)
- let cse_var_56: int32 = (cse_var_2 + 193)
- let cse_var_55: int32 = (cse_var_2 + 192)
- let cse_var_54: int32 = (cse_var_2 + 175)
- let cse_var_53: int32 = (cse_var_2 + 174)
- let cse_var_52: int32 = (cse_var_2 + 173)
- let cse_var_51: int32 = (cse_var_2 + 172)
- let cse_var_50: int32 = (cse_var_2 + 171)
- let cse_var_49: int32 = (cse_var_2 + 170)
- let cse_var_48: int32 = (cse_var_2 + 169)
- let cse_var_47: int32 = (cse_var_2 + 168)
- let cse_var_46: int32 = (cse_var_2 + 167)
- let cse_var_45: int32 = (cse_var_2 + 166)
- let cse_var_44: int32 = (cse_var_2 + 165)
- let cse_var_43: int32 = (cse_var_2 + 164)
- let cse_var_42: int32 = (cse_var_2 + 163)
- let cse_var_41: int32 = (cse_var_2 + 162)
- let cse_var_40: int32 = (cse_var_2 + 161)
- let cse_var_39: int32 = (cse_var_2 + 160)
- let cse_var_38: int32 = (cse_var_2 + 15)
- let cse_var_37: int32 = (cse_var_2 + 143)
- let cse_var_36: int32 = (cse_var_2 + 142)
- let cse_var_35: int32 = (cse_var_2 + 141)
- let cse_var_34: int32 = (cse_var_2 + 140)
- let cse_var_33: int32 = (cse_var_2 + 14)
- let cse_var_32: int32 = (cse_var_2 + 139)
- let cse_var_31: int32 = (cse_var_2 + 138)
- let cse_var_30: int32 = (cse_var_2 + 137)
- let cse_var_29: int32 = (cse_var_2 + 136)
- let cse_var_28: int32 = (cse_var_2 + 135)
- let cse_var_27: int32 = (cse_var_2 + 134)
- let cse_var_26: int32 = (cse_var_2 + 133)
- let cse_var_25: int32 = (cse_var_2 + 132)
- let cse_var_24: int32 = (cse_var_2 + 131)
- let cse_var_23: int32 = (cse_var_2 + 130)
- let cse_var_22: int32 = (cse_var_2 + 13)
- let cse_var_21: int32 = (cse_var_2 + 129)
- let cse_var_20: int32 = (cse_var_2 + 128)
- let cse_var_19: int32 = (cse_var_2 + 12)
- let cse_var_18: int32 = (cse_var_2 + 111)
- let cse_var_17: int32 = (cse_var_2 + 110)
- let cse_var_16: int32 = (cse_var_2 + 11)
- let cse_var_15: int32 = (cse_var_2 + 109)
- let cse_var_14: int32 = (cse_var_2 + 108)
- let cse_var_13: int32 = (cse_var_2 + 107)
- let cse_var_12: int32 = (cse_var_2 + 106)
- let cse_var_11: int32 = (cse_var_2 + 105)
- let cse_var_10: int32 = (cse_var_2 + 104)
- let cse_var_9: int32 = (cse_var_2 + 103)
- let cse_var_8: int32 = (cse_var_2 + 102)
- let cse_var_7: int32 = (cse_var_2 + 101)
- let cse_var_6: int32 = (cse_var_2 + 100)
- let cse_var_5: int32 = (cse_var_2 + 10)
- let cse_var_4: int32 = (cse_var_2 + 1)
- let cse_var_3: int32 = (floordiv(i0.outer.i1.outer.fused, 16)*2048)
- {
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_63] = (compute_5[cse_var_63] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_88] = (compute_5[cse_var_88] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_97] = (compute_5[cse_var_97] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_106] = (compute_5[cse_var_106] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_107] = (compute_5[cse_var_107] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_114] = (compute_5[cse_var_114] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_125] = (compute_5[cse_var_125] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_126] = (compute_5[cse_var_126] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_22] = (compute_5[cse_var_22] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_33] = (compute_5[cse_var_33] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_38] = (compute_5[cse_var_38] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_89] = (compute_5[cse_var_89] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_90] = (compute_5[cse_var_90] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_91] = (compute_5[cse_var_91] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_92] = (compute_5[cse_var_92] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_93] = (compute_5[cse_var_93] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_94] = (compute_5[cse_var_94] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_95] = (compute_5[cse_var_95] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_96] = (compute_5[cse_var_96] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_98] = (compute_5[cse_var_98] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_99] = (compute_5[cse_var_99] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_100] = (compute_5[cse_var_100] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_101] = (compute_5[cse_var_101] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_102] = (compute_5[cse_var_102] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_103] = (compute_5[cse_var_103] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_104] = (compute_5[cse_var_104] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_105] = (compute_5[cse_var_105] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_108] = (compute_5[cse_var_108] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_109] = (compute_5[cse_var_109] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_110] = (compute_5[cse_var_110] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_111] = (compute_5[cse_var_111] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_112] = (compute_5[cse_var_112] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_113] = (compute_5[cse_var_113] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_115] = (compute_5[cse_var_115] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_116] = (compute_5[cse_var_116] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_117] = (compute_5[cse_var_117] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_118] = (compute_5[cse_var_118] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_119] = (compute_5[cse_var_119] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_120] = (compute_5[cse_var_120] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_121] = (compute_5[cse_var_121] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_122] = (compute_5[cse_var_122] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_123] = (compute_5[cse_var_123] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_124] = (compute_5[cse_var_124] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_127] = (compute_5[cse_var_127] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_128] = (compute_5[cse_var_128] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_129] = (compute_5[cse_var_129] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_130] = (compute_5[cse_var_130] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_18] = (compute_5[cse_var_18] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_20] = (compute_5[cse_var_20] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_21] = (compute_5[cse_var_21] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_23] = (compute_5[cse_var_23] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_24] = (compute_5[cse_var_24] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_25] = (compute_5[cse_var_25] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_26] = (compute_5[cse_var_26] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_27] = (compute_5[cse_var_27] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_28] = (compute_5[cse_var_28] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_29] = (compute_5[cse_var_29] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_30] = (compute_5[cse_var_30] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_31] = (compute_5[cse_var_31] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_32] = (compute_5[cse_var_32] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_34] = (compute_5[cse_var_34] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_35] = (compute_5[cse_var_35] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_36] = (compute_5[cse_var_36] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_37] = (compute_5[cse_var_37] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
- compute_5[cse_var_39] = (compute_5[cse_var_39] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_40] = (compute_5[cse_var_40] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_41] = (compute_5[cse_var_41] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_42] = (compute_5[cse_var_42] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_43] = (compute_5[cse_var_43] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_44] = (compute_5[cse_var_44] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_45] = (compute_5[cse_var_45] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_46] = (compute_5[cse_var_46] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_47] = (compute_5[cse_var_47] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_48] = (compute_5[cse_var_48] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_49] = (compute_5[cse_var_49] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_50] = (compute_5[cse_var_50] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_51] = (compute_5[cse_var_51] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_52] = (compute_5[cse_var_52] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_53] = (compute_5[cse_var_53] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_54] = (compute_5[cse_var_54] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
- compute_5[cse_var_55] = (compute_5[cse_var_55] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_56] = (compute_5[cse_var_56] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_57] = (compute_5[cse_var_57] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_58] = (compute_5[cse_var_58] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_59] = (compute_5[cse_var_59] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_60] = (compute_5[cse_var_60] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_61] = (compute_5[cse_var_61] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_62] = (compute_5[cse_var_62] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_64] = (compute_5[cse_var_64] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_65] = (compute_5[cse_var_65] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_66] = (compute_5[cse_var_66] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_67] = (compute_5[cse_var_67] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_68] = (compute_5[cse_var_68] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_69] = (compute_5[cse_var_69] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_70] = (compute_5[cse_var_70] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_71] = (compute_5[cse_var_71] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
- compute_5[cse_var_72] = (compute_5[cse_var_72] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_131)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_73] = (compute_5[cse_var_73] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_74] = (compute_5[cse_var_74] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_75] = (compute_5[cse_var_75] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_76] = (compute_5[cse_var_76] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_77] = (compute_5[cse_var_77] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_78] = (compute_5[cse_var_78] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_79] = (compute_5[cse_var_79] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_80] = (compute_5[cse_var_80] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_81] = (compute_5[cse_var_81] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_82] = (compute_5[cse_var_82] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_83] = (compute_5[cse_var_83] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_84] = (compute_5[cse_var_84] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_85] = (compute_5[cse_var_85] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_86] = (compute_5[cse_var_86] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
- compute_5[cse_var_87] = (compute_5[cse_var_87] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_131) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ preflattened_buffer_map = {placeholder_6: placeholder_15: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_7: placeholder_16: Buffer(placeholder_12, int32, [4916], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_18: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_19: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 32) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [2048]), storage_scope = global {
+ for (i.outer.inner: int32, 0, 64) {
+ for (i.inner.init: int32, 0, 2) {
+ let cse_var_1: int32 = ((i.outer.inner*32) + (i.inner.init*16))
+ {
+ compute_5: Buffer(compute_4, float32, [2048], [])[cse_var_1] = 0f32
+ compute_5[(cse_var_1 + 1)] = 0f32
+ compute_5[(cse_var_1 + 2)] = 0f32
+ compute_5[(cse_var_1 + 3)] = 0f32
+ compute_5[(cse_var_1 + 4)] = 0f32
+ compute_5[(cse_var_1 + 5)] = 0f32
+ compute_5[(cse_var_1 + 6)] = 0f32
+ compute_5[(cse_var_1 + 7)] = 0f32
+ compute_5[(cse_var_1 + 8)] = 0f32
+ compute_5[(cse_var_1 + 9)] = 0f32
+ compute_5[(cse_var_1 + 10)] = 0f32
+ compute_5[(cse_var_1 + 11)] = 0f32
+ compute_5[(cse_var_1 + 12)] = 0f32
+ compute_5[(cse_var_1 + 13)] = 0f32
+ compute_5[(cse_var_1 + 14)] = 0f32
+ compute_5[(cse_var_1 + 15)] = 0f32
+ }
+ }
+ for (elem_idx: int32, 0, (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])) {
+ for (i.inner: int32, 0, 2) {
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_2: int32 = ((i.outer.inner*32) + (i.inner*16))
+ compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16))]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_3: int32 = (((i.outer.inner*32) + (i.inner*16)) + 1)
+ compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_4: int32 = (((i.outer.inner*32) + (i.inner*16)) + 2)
+ compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_5: int32 = (((i.outer.inner*32) + (i.inner*16)) + 3)
+ compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_6: int32 = (((i.outer.inner*32) + (i.inner*16)) + 4)
+ compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_7: int32 = (((i.outer.inner*32) + (i.inner*16)) + 5)
+ compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_8: int32 = (((i.outer.inner*32) + (i.inner*16)) + 6)
+ compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_9: int32 = (((i.outer.inner*32) + (i.inner*16)) + 7)
+ compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_10: int32 = (((i.outer.inner*32) + (i.inner*16)) + 8)
+ compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_11: int32 = (((i.outer.inner*32) + (i.inner*16)) + 9)
+ compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_12: int32 = (((i.outer.inner*32) + (i.inner*16)) + 10)
+ compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_13: int32 = (((i.outer.inner*32) + (i.inner*16)) + 11)
+ compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_14: int32 = (((i.outer.inner*32) + (i.inner*16)) + 12)
+ compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_15: int32 = (((i.outer.inner*32) + (i.inner*16)) + 13)
+ compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_16: int32 = (((i.outer.inner*32) + (i.inner*16)) + 14)
+ compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
+ let cse_var_17: int32 = (((i.outer.inner*32) + (i.inner*16)) + 15)
+ compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((i.outer.inner*512) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
}
}
}
}
- for (i0.inner: int32, 0, 8) {
- let cse_var_132: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*4096) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
- compute[ramp(cse_var_132, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_132, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 128) {
+ let cse_var_18: int32 = ((i0.inner*512) + (i0.outer.i1.outer.fused*16))
+ compute[ramp(cse_var_18, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_18, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -1058,7 +754,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 2.694 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 3.632 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 1b3fa9834..6f9e34ac3 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -322,7 +322,7 @@
<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>00:46.715</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:46.861</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -331,11 +331,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:46.678</p></td>
+<td><p>00:46.825</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:00.021</p></td>
+<td><p>00:00.020</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
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 d23245169..bd1b92f7c 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -1431,8 +1431,8 @@ No: 8 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
TimeoutError
[('tile_f', [-1, 2, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4909501
-No: 9 GFLOPS: 173.43/173.43 result: MeasureResult(costs=(0.0013348717333333334,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8318445682525635, timestamp=1658819800.0006714) [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5072689
-No: 10 GFLOPS: 0.00/173.43 result: Traceback (most recent call last):
+No: 9 GFLOPS: 122.21/122.21 result: MeasureResult(costs=(0.0018942681607142856,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0902655124664307, timestamp=1658826629.7328916) [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5072689
+No: 10 GFLOPS: 0.00/122.21 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1555,8 +1555,8 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5092711
-No: 11 GFLOPS: 261.23/261.23 result: MeasureResult(costs=(0.0008862034751381216,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7316110134124756, timestamp=1658819800.9196448) [('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
-No: 12 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+No: 11 GFLOPS: 261.55/261.55 result: MeasureResult(costs=(0.0008851145303867403,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7998321056365967, timestamp=1658826630.6496062) [('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
+No: 12 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1679,7 +1679,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 128, 1, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,183542
-No: 13 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+No: 13 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1802,7 +1802,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 8, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2482196
-No: 14 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+No: 14 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1925,9 +1925,9 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10306226
-No: 15 GFLOPS: 5.48/261.23 result: MeasureResult(costs=(0.04226059,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.826474666595459, timestamp=1658819805.474508) [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5330964
-No: 16 GFLOPS: 3.36/261.23 result: MeasureResult(costs=(0.06892332475,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.549631118774414, timestamp=1658819806.7095149) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2140058
-No: 17 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+No: 15 GFLOPS: 5.39/261.55 result: MeasureResult(costs=(0.042985643500000004,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9060659408569336, timestamp=1658826635.3068864) [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5330964
+No: 16 GFLOPS: 3.35/261.55 result: MeasureResult(costs=(0.06907856475,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.6305437088012695, timestamp=1658826636.5465927) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2140058
+No: 17 GFLOPS: 0.00/261.55 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/python3.7/concurrent/futures/_base.py", line 435, in result
@@ -1945,8 +1945,8 @@ No: 17 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
TimeoutError
[('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10195251
-No: 18 GFLOPS: 27.15/261.23 result: MeasureResult(costs=(0.0085280948125,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3287360668182373, timestamp=1658819817.7805817) [('tile_f', [-1, 4, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6068603
-No: 19 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+No: 18 GFLOPS: 28.14/261.55 result: MeasureResult(costs=(0.008227821785714285,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.295943260192871, timestamp=1658826647.5799015) [('tile_f', [-1, 4, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6068603
+No: 19 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2069,7 +2069,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, 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, 16, 4, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6956993
-No: 20 GFLOPS: 0.00/261.23 result: Traceback (most recent call last):
+No: 20 GFLOPS: 0.00/261.55 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2232,7 +2232,7 @@ and measure running time.</p>
Best config:
[('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
Finish loading 20 records
-Time cost of this operator: 0.001217
+Time cost of this operator: 0.001272
</pre></div>
</div>
<div class="sphx-glr-footer 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 b242b13eb..885837eec 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -578,10 +578,10 @@ the tuned operator.</p>
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 311.1 98.734 (1, 2, 10, 10, 3) 2 1 [311.1]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.035 0.963 (1, 6, 10, 10) 1 1 [3.035]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.954 0.303 (1, 1, 10, 10, 3) 1 1 [0.954]
-Total_time - 315.089 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.2 98.732 (1, 2, 10, 10, 3) 2 1 [312.2]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.03 0.958 (1, 6, 10, 10) 1 1 [3.03]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.98 0.31 (1, 1, 10, 10, 3) 1 1 [0.98]
+Total_time - 316.21 - - - - -
</pre></div>
</div>
</div>
@@ -634,10 +634,10 @@ Total_time -
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 79.688 96.638 (1, 6, 10, 10, 1) 2 1 [79.688]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.802 2.185 (1, 6, 10, 10) 1 1 [1.802]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.97 1.177 (1, 1, 10, 10, 3) 1 1 [0.97]
-Total_time - 82.459 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 79.625 96.391 (1, 6, 10, 10, 1) 2 1 [79.625]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 2.012 2.436 (1, 6, 10, 10) 1 1 [2.012]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.969 1.173 (1, 1, 10, 10, 3) 1 1 [0.969]
+Total_time - 82.606 - - - - -
</pre></div>
</div>
<div class="sphx-glr-footer 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_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index 06292cc74..b5c1662a1 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -510,7 +510,7 @@ take about <strong>2 minutes</strong> to download the Stanford Cars, while COCO
<a href="https://docs.python.org/3/library/shutil.html#shutil.move" title="shutil.move" class="sphx-glr-backref-module-shutil sphx-glr-backref-type-py-function"><span class="n">shutil</span><span class="o">.</span><span class="n">move</span></a><span class="p">(</span><span class="sa">f</span><span class="s2">"</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-typ [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmpfld0ithc/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmpdi_m8g8c/images/random'
</pre></div>
</div>
</div>
@@ -570,8 +570,8 @@ objects to other stuff? We can display some examples from our datasets using <co
<span class="n">plt</span><span class="o">.</span><span class="n">axis</span><span class="p">(</span><span class="s2">"off"</span><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpfld0ithc/images/target contains 8144 images
-/tmp/tmpfld0ithc/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpdi_m8g8c/images/target contains 8144 images
+/tmp/tmpdi_m8g8c/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -683,13 +683,13 @@ the time on our validation set).</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Epoch 1/3
-328/328 - 55s - loss: 0.2105 - accuracy: 0.9265 - val_loss: 0.1443 - val_accuracy: 0.9577
+328/328 - 55s - loss: 0.2380 - accuracy: 0.9218 - val_loss: 0.1577 - val_accuracy: 0.9520
Epoch 2/3
-328/328 - 52s - loss: 0.0941 - accuracy: 0.9628 - val_loss: 0.1303 - val_accuracy: 0.9622
+328/328 - 53s - loss: 0.1005 - accuracy: 0.9638 - val_loss: 0.1300 - val_accuracy: 0.9592
Epoch 3/3
-328/328 - 52s - loss: 0.0649 - accuracy: 0.9778 - val_loss: 0.1483 - val_accuracy: 0.9551
+328/328 - 53s - loss: 0.0670 - accuracy: 0.9748 - val_loss: 0.1283 - val_accuracy: 0.9641
-<keras.callbacks.History object at 0x7f0081213590>
+<keras.callbacks.History object at 0x7fd5c8c90d50>
</pre></div>
</div>
</div>
@@ -951,7 +951,7 @@ as intended.</p>
<p>From here, we could modify the model to read live images from the camera - we have another
Arduino tutorial for how to do that <a class="reference external" href="https://github.com/guberti/tvm-arduino-demos/tree/master/examples/person_detection">on GitHub</a>. Alternatively, we could also
<a class="reference external" href="https://tvm.apache.org/docs/how_to/work_with_microtvm/micro_autotune.html">use TVM’s autotuning capabilities</a> to dramatically improve the model’s performance.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 5 minutes 26.769 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 5 minutes 14.789 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-train-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/b52cec46baf4f78d6bcd94cbe269c8a6/micro_train.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">micro_train.py</span></code></a></p>
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 39ccf0788..5500d694b 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -322,7 +322,7 @@
<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>06:12.487</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>06:03.069</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -331,15 +331,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_train.html#sphx-glr-how-to-work-with-microtvm-micro-train-py"><span class="std std-ref">Training Vision Models for microTVM on Arduino</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_train.py</span></code>)</p></td>
-<td><p>05:26.769</p></td>
+<td><p>05:14.789</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:42.434</p></td>
+<td><p>00:44.804</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:03.282</p></td>
+<td><p>00:03.474</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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 with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></td>
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 bd2cd6484..51a33bd7a 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -322,7 +322,7 @@
<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>00:41.507</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:43.332</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -331,15 +331,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="using_pipeline_executor.html#sphx-glr-how-to-work-with-relay-using-pipeline-executor-py"><span class="std std-ref">Using Pipeline Executor in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_pipeline_executor.py</span></code>)</p></td>
-<td><p>00:30.038</p></td>
+<td><p>00:31.637</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:09.706</p></td>
+<td><p>00:09.944</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:01.756</p></td>
+<td><p>00:01.745</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
diff --git a/docs/how_to/work_with_schedules/intrin_math.html b/docs/how_to/work_with_schedules/intrin_math.html
index 3f83e27ad..341f0512b 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -517,7 +517,7 @@ The following example customizes CUDA lowering rule for <code class="code docuti
<a href="../../reference/api/python/ir.html#tvm.ir.register_intrin_lowering" title="tvm.ir.register_intrin_lowering" class="sphx-glr-backref-module-tvm-ir sphx-glr-backref-type-py-function"><span class="n">register_intrin_lowering</span></a><span class="p">(</span><span class="s2">"tir.exp"</span><span class="p">,</span> <span class="n">target</span><span class="o">=</span><span class="s2">"cuda"</span><span class="p">,</span> <span class="n">f</span><span class="o">= [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7efff8247710>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7fd5d82b88c0>
</pre></div>
</div>
<p>Register the rule to TVM with override option to override existing rule.
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 a23e0f9c5..837070250 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -322,7 +322,7 @@
<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>00:04.108</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:04.412</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -331,35 +331,35 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:01.907</p></td>
+<td><p>00:02.085</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:00.974</p></td>
+<td><p>00:01.001</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:00.532</p></td>
+<td><p>00:00.578</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:00.517</p></td>
+<td><p>00:00.561</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:00.097</p></td>
+<td><p>00:00.103</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:00.040</p></td>
+<td><p>00:00.042</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><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></td>
-<td><p>00:00.026</p></td>
+<td><p>00:00.027</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><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></td>
-<td><p>00:00.014</p></td>
+<td><p>00:00.015</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index b6f63a269..7153fe400 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -572,7 +572,7 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpdk5__7ic/input0.cc'\nsource_filename = \"/tmp/tmpdk5__7ic/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpavq6yxme/input0.cc'\nsource_filename = \"/tmp/tmpavq6yxme/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index ba99d4576..bc39f8b59 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1597,7 +1597,7 @@ history states as starting point to perform Evolutionary Search).</p></li>
<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>
@@ -1881,7 +1881,7 @@ Candidates:
<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">
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index dbcdb8bdc..427d48588 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/9bef7de9f/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/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/9bef7de9f/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/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/9bef7de9f/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/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/9bef7de9f/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/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/9bef7de9f/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/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">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/cachedcallstack.html b/docs/reference/api/typedoc/classes/cachedcallstack.html
index 44151550b..1ab26e8cb 100644
--- a/docs/reference/api/typedoc/classes/cachedcallstack.html
+++ b/docs/reference/api/typedoc/classes/cachedcallstack.html
@@ -144,7 +144,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L223">memory.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L223">memory.ts:223</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -172,7 +172,7 @@
<div class="tsd-signature tsd-kind-icon">temp<wbr>Args<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><a href="../interfaces/disposable.html" class="tsd-signature-type">Disposable</a><span class="tsd-signature-symbol">></span><span class="tsd-signature-symbol"> = []</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L208">memory.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L208">memory.ts:208</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -194,7 +194,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L312">memory.ts:312</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L312">memory.ts:312</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -226,7 +226,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L284">memory.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L284">memory.ts:284</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -262,7 +262,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L388">memory.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L388">memory.ts:388</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -300,7 +300,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L376">memory.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L376">memory.ts:376</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -340,7 +340,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L267">memory.ts:267</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L267">memory.ts:267</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -373,7 +373,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L243">memory.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L243">memory.ts:243</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -390,7 +390,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L321">memory.ts:321</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L321">memory.ts:321</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -422,7 +422,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L252">memory.ts:252</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L252">memory.ts:252</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -444,7 +444,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L359">memory.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L359">memory.ts:359</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -470,7 +470,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/9bef7de9f/web/src/memory.ts#L342">memory.ts:342</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ea6ea4275/web/src/memory.ts#L342">memory.ts:342</a></li>
</ul>
... 2751 lines suppressed ...