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 2023/05/05 10:32:57 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@eca6edf7faf8d8ec0a6ea09b5d6f1ba06153a594)
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 f336dac736 deploying docs (apache/tvm@eca6edf7faf8d8ec0a6ea09b5d6f1ba06153a594)
f336dac736 is described below
commit f336dac7363d585a22e3de2d525990206d57b64a
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Fri May 5 10:32:51 2023 +0000
deploying docs (apache/tvm@eca6edf7faf8d8ec0a6ea09b5d6f1ba06153a594)
---
docs/_images/sphx_glr_micro_train_001.png | Bin 327396 -> 344888 bytes
docs/_images/sphx_glr_micro_train_thumb.png | Bin 23429 -> 24077 bytes
.../how_to/compile_models/from_darknet.rst.txt | 2 +-
.../how_to/compile_models/from_keras.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_paddle.rst.txt | 5 -
.../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_adreno.rst.txt | 4 +-
.../deploy_model_on_adreno_tvmc.rst.txt | 2 +-
.../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 | 24 +-
.../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 | 1916 +++++++++++---------
.../tune_network_cuda.rst.txt | 4 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 411 ++++-
.../tune_with_autotvm/sg_execution_times.rst.txt | 10 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 110 +-
.../work_with_microtvm/micro_autotune.rst.txt | 18 +-
.../work_with_microtvm/micro_pytorch.rst.txt | 4 +-
.../how_to/work_with_microtvm/micro_train.rst.txt | 18 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 14 +-
.../work_with_relay/sg_execution_times.rst.txt | 10 +-
.../how_to/work_with_schedules/intrin_math.rst.txt | 2 +-
.../work_with_schedules/sg_execution_times.rst.txt | 18 +-
.../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 | 4 +-
docs/_sources/tutorial/autotvm_matmul_x86.rst.txt | 20 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 57 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 18 +-
.../tutorial/tensor_expr_get_started.rst.txt | 44 +-
docs/commit_hash | 2 +-
docs/how_to/compile_models/from_darknet.html | 2 +-
docs/how_to/compile_models/from_keras.html | 2 +-
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_oneflow.html | 14 +-
docs/how_to/compile_models/from_paddle.html | 1 -
docs/how_to/compile_models/from_pytorch.html | 14 +-
docs/how_to/compile_models/from_tensorflow.html | 2 +-
docs/how_to/compile_models/sg_execution_times.html | 22 +-
.../deploy_models/deploy_model_on_adreno.html | 4 +-
.../deploy_models/deploy_model_on_adreno_tvmc.html | 31 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 58 +-
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 | 36 +-
docs/how_to/deploy_models/sg_execution_times.html | 24 +-
.../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 | 1911 ++++++++++---------
.../tune_with_autoscheduler/tune_network_cuda.html | 4 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 411 ++++-
.../tune_with_autotvm/sg_execution_times.html | 10 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 110 +-
docs/how_to/work_with_microtvm/micro_autotune.html | 18 +-
docs/how_to/work_with_microtvm/micro_pytorch.html | 4 +-
docs/how_to/work_with_microtvm/micro_train.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 14 +-
.../how_to/work_with_relay/sg_execution_times.html | 10 +-
docs/how_to/work_with_schedules/intrin_math.html | 2 +-
.../work_with_schedules/sg_execution_times.html | 18 +-
docs/install/nnpack.html | 12 +-
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 +-
docs/reference/api/typedoc/classes/instance.html | 58 +-
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 +-
.../api/typedoc/classes/runtimecontext.html | 22 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
docs/reference/api/typedoc/classes/tvmarray.html | 16 +-
docs/reference/api/typedoc/classes/tvmobject.html | 12 +-
.../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 | 124 +-
.../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 | 4 +-
docs/tutorial/autotvm_matmul_x86.html | 20 +-
docs/tutorial/autotvm_relay_x86.html | 261 ++-
docs/tutorial/cross_compilation_and_rpc.html | 2 +-
docs/tutorial/intro_topi.html | 2 +-
docs/tutorial/sg_execution_times.html | 22 +-
docs/tutorial/tensor_expr_get_started.html | 44 +-
134 files changed, 3796 insertions(+), 2833 deletions(-)
diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index 6a9ff83b6f..03eff51625 100644
Binary files a/docs/_images/sphx_glr_micro_train_001.png and b/docs/_images/sphx_glr_micro_train_001.png differ
diff --git a/docs/_images/sphx_glr_micro_train_thumb.png b/docs/_images/sphx_glr_micro_train_thumb.png
index faf94be3ac..ce827177eb 100644
Binary files a/docs/_images/sphx_glr_micro_train_thumb.png and b/docs/_images/sphx_glr_micro_train_thumb.png differ
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 dd1a5ed813..1de4d87ae8 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -318,7 +318,7 @@ The process is no different from other examples.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 28.330 seconds)
+ **Total running time of the script:** ( 1 minutes 24.945 seconds)
.. _sphx_glr_download_how_to_compile_models_from_darknet.py:
diff --git a/docs/_sources/how_to/compile_models/from_keras.rst.txt b/docs/_sources/how_to/compile_models/from_keras.rst.txt
index 94fb2d025d..91bdc62360 100644
--- a/docs/_sources/how_to/compile_models/from_keras.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_keras.rst.txt
@@ -232,7 +232,7 @@ Look up prediction top 1 index in 1000 class synset.
.. code-block:: none
Relay top-1 id: 285, class name: Egyptian cat
-
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 2s 2s/step
+
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 1s 1s/step
Keras top-1 id: 285, class name: Egyptian cat
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 05c0ac3e74..70a1146de0 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -116,7 +116,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip3bdf3f20-1bc9-4501-aa3a-7b59d9a0477e from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip1efd7c6f-5a68-4c9c-96e2-111990adc926 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 ce5a914cb7..f875be2f3a 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -121,7 +121,7 @@ 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, 52.9MB/s]
39%|###8 | 16.0M/41.5M [00:00<00:00, 52.3MB/s]
58%|#####7 | 24.0M/41.5M [00:00<00:00, 58.6MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 57.9MB/s]
91%|######### | 37.6M/41.5M [00:00<00:00, 49.2MB/s]
100%|##########| 41.5M/41.5M [00:01<00:00, 38.0MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
15%|#5 | 6.33M/41.5M [00:00<00:00, 48.1MB/s]
26%|##6 | 10.9M/41.5M [00:00<00:00, 44.8MB/s]
37%|###6 | 15.2M/41.5M [00:00<00:00, 34.3MB/s]
45%|####4 | 18.6M/41.5M [00:00<00:00, 33.5MB/s]
63%|######2 | 26.1M/41.5M [00:00<00:00, 46.9MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 46.4MB/s]
92%|#########2| 38.3M/41.5M [00:00<00:00, 41.8MB/s]
100%|##########| 41.5M/41.5M [00:01<00:00, 41.1MB/s]
diff --git a/docs/_sources/how_to/compile_models/from_paddle.rst.txt b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
index 37a75a81bd..8df75c9e33 100644
--- a/docs/_sources/how_to/compile_models/from_paddle.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
@@ -207,11 +207,6 @@ Look up prediction top 1 index in 1000 class synset.
-.. rst-class:: sphx-glr-timing
-
- **Total running time of the script:** ( 1 minutes 4.701 seconds)
-
-
.. _sphx_glr_download_how_to_compile_models_from_paddle.py:
.. only:: html
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 1f58b0b1b9..4a572056c5 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -101,7 +101,7 @@ Load a pretrained PyTorch model
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
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]
14%|#4 | 6.30M/44.7M [00:00<00:00, 56.2MB/s]
26%|##6 | 11.7M/44.7M [00:00<00:00, 42.4MB/s]
36%|###5 | 16.0M/44.7M [00:00<00:00, 40.6MB/s]
58%|#####8 | 26.0M/44.7M [00:00<00:00, 61.8MB/s]
79%|#######9 | 35.4M/44.7M [00:00<00:00, 73.6MB/s]
99%|#########8| 44.2M/44.7M [00:00<00:00, 78.9MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 66.8MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
18%|#7 | 7.99M/44.7M [00:00<00:00, 59.8MB/s]
36%|###5 | 16.0M/44.7M [00:00<00:00, 57.5MB/s]
54%|#####3 | 24.0M/44.7M [00:00<00:00, 63.9MB/s]
68%|######7 | 30.2M/44.7M [00:00<00:00, 57.1MB/s]
80%|#######9 | 35.7M/44.7M [00:00<00:00, 48.9MB/s]
91%|######### | 40.5M/44.7M [00:00<00:00, 45.9MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 55.5MB/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 9d0bb66886..fd318aa09b 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -430,7 +430,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 40.711 seconds)
+ **Total running time of the script:** ( 1 minutes 31.607 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 8191ba79f2..c229bed4c5 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
=================
-**07:35.835** total execution time for **how_to_compile_models** files:
+**07:03.903** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:40.711 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:31.607 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:28.330 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:24.945 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 01:04.701 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:59.391 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:41.806 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:38.869 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:37.252 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:34.177 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:33.273 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:31.393 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:30.294 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:29.298 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:29.209 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:26.631 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:27.269 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:24.751 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.991 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.841 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
index eff49257ed..ed029d3278 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
@@ -673,7 +673,7 @@ well as provides information about the model's performance
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 3354.0283 3351.9853 3365.8622 3349.3537 4.8277
+ 3345.4809 3344.7137 3354.2131 3340.3818 3.6215
@@ -682,7 +682,7 @@ well as provides information about the model's performance
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 8.105 seconds)
+ **Total running time of the script:** ( 1 minutes 5.981 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_model_on_adreno.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
index 463673d7d3..5a199a39f8 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
@@ -127,7 +127,7 @@ Make a Keras Resnet50 Model
.. code-block:: none
Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5
-
8192/102967424 [..............................] - ETA: 0s
6635520/102967424 [>.............................] - ETA: 0s
8380416/102967424 [=>............................] - ETA: 1s
15024128/102967424 [===>..........................] - ETA: 1s
16769024/102967424 [===>..........................] - ETA: 1s
23412736/102967424 [=====>........................] - ETA: 1s
25157632/102967424 [======>.......................] - ETA: 1s
33546240/102967424 [========>.....................] - ETA: 1s
41934848/102967424 [===========>..................] - ETA: 1s
48578560/102967424 [=============>................] - ETA: 1s
50323456/102967424 [=============>................] - ETA: 1s
56967168/102967424 [===============>..............] - ETA: 0s
58712064/102967424 [================>.............] - ETA: 1s
65355776/102967424 [==================>...........] - ETA: 0s
67100672/102967424 [==================>...........] - ETA: 0s
69296128/102967424 [===================>..........] -
ETA: 0s
75202560/102967424 [====================>.........] - ETA: 0s
75489280/102967424 [====================>.........] - ETA: 0s
76972032/102967424 [=====================>........] - ETA: 0s
82124800/102967424 [======================>.......] - ETA: 0s
83877888/102967424 [=======================>......] - ETA: 0s
90521600/102967424 [=========================>....] - ETA: 0s
92266496/102967424 [=========================>....] - ETA: 0s
98910208/102967424
[===========================>..] - ETA: 0s
100646912/102967424 [============================>.] - ETA: 0s
102850560/102967424 [============================>.] - ETA: 0s
102967424/102967424 [==============================] - 3s 0us/step
+
8192/102967424 [..............................] - ETA: 0s
8380416/102967424 [=>............................] - ETA: 0s
16769024/102967424 [===>..........................] - ETA: 0s
19562496/102967424 [====>.........................] - ETA: 0s
25157632/102967424 [======>.......................] - ETA: 1s
27992064/102967424 [=======>......................] - ETA: 1s
33546240/102967424 [========>.....................] - ETA: 1s
35741696/102967424 [=========>....................] - ETA: 1s
41934848/102967424 [===========>..................] - ETA: 0s
42065920/102967424 [===========>..................] - ETA: 1s
50323456/102967424 [=============>................] - ETA: 0s
58712064/102967424 [================>.............] - ETA: 0s
60932096/102967424 [================>.............] - ETA: 0s
67100672/102967424 [==================>...........] - ETA: 0s
69296128/102967424 [===================>..........] - ETA: 0s
75202560/102967424 [====================>.........] -
ETA: 0s
76529664/102967424 [=====================>........] - ETA: 0s
82124800/102967424 [======================>.......] - ETA: 0s
84926464/102967424 [=======================>......] - ETA: 0s
90521600/102967424 [=========================>....] - ETA: 0s
92266496/102967424 [=========================>....] - ETA: 0s
94461952/102967424 [==========================>...] - ETA: 0s
102850560/102967424 [============================>.] - ETA: 0s
102967424/102967424
[==============================] - 2s 0us/step
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 69f6759f10..3bee55a2b5 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
@@ -437,7 +437,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 17.2645 17.0776 17.8742 16.7083 0.4020
+ 16.2511 16.0867 17.7078 15.9810 0.4937
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 2834b685d7..4f3942c8c6 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
@@ -130,7 +130,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MaskRCNN_ResNet50_FPN_Weights.COCO_V1`. You can also use `weights=MaskRCNN_ResNet50_FPN_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
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]
5%|4 | 7.99M/170M [00:00<00:02, 56.6MB/s]
9%|8 | 14.5M/170M [00:00<00:02, 62.5MB/s]
12%|#2 | 20.6M/170M [00:00<00:02, 63.0MB/s]
16%|#5 | 26.7M/170M [00:00<00:04, 32.7MB/s]
19%|#8 | 32.0M/170M [00:00<00:04, 30.9MB/s]
24%|##3 | 40.0M/170M [00:01<00:03, 39.1MB/s]
28%|##8 | 48.0M/170M [00:01<00:03, 42.4MB/s]
33%|###2 | 56.0M/170M [00:01<00:02, 48.7MB/s]
38%|###7 | 64.0M/170M [00:01<00:01, 55.7MB/s]
42%|####2 | 72.0M/170M [00:01<00:01, 52.4MB/s]
46%|####6 | 78.3M/170M [00:01<00:02, 46.4MB/s]
49%|####8 | 83.1M/170M [00:02<00:02, 38.2MB/s]
52%|#####1 | 88.0M/170M [00:02<00:02, 36.8MB/s]
55%|#####4 | 93.2M/170M [00:02<00:02, 38.2MB/s]
57%|#####7 | 97.0M/170M [00:02<00:02, 35.9MB/s]
61%|######1 | 104M/170M [00:02<00:01, 43.3MB/s]
64%|######3 | 108M/170M [00:02<00:01, 42.8MB/s
]
66%|######6 | 112M/170M [00:02<00:01, 40.1MB/s]
71%|####### | 120M/170M [00:02<00:01, 44.5MB/s]
74%|#######4 | 126M/170M [00:03<00:01, 45.2MB/s]
79%|#######9 | 134M/170M [00:03<00:00, 47.6MB/s]
82%|########1 | 139M/170M [00:03<00:00, 47.4MB/s]
85%|########4 | 144M/170M [00:03<00:00, 43.1MB/s]
88%|########8 | 150M/170M [00:03<00:00, 44.2MB/s]
91%|#########1| 155M/170M [00:03<00:00, 39.0MB/s]
93%|#########3| 158M/170M [00:04<00:00, 29.5MB/s]
95%|#########5| 161M/170M [00:04<00:00, 26.4MB/s]
98%|#########7| 166M/170M [00:04<00:00, 29.8MB/s]
100%|##########| 170M/170M [00:04<00:00, 40.3MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
5%|4 | 7.99M/170M [00:00<00:03, 42.9MB/s]
8%|8 | 14.3M/170M [00:00<00:03, 48.6MB/s]
11%|#1 | 19.1M/170M [00:00<00:03, 49.0MB/s]
14%|#4 | 23.9M/170M [00:00<00:03, 39.1MB/s]
16%|#6 | 27.8M/170M [00:00<00:04, 36.8MB/s]
19%|#8 | 31.8M/170M [00:00<00:03, 38.2MB/s]
21%|## | 35.6M/170M [00:00<00:03, 35.2MB/s]
24%|##3 | 40.0M/170M [00:01<00:03, 36.6MB/s]
28%|##8 | 48.0M/170M [00:01<00:03, 39.2MB/s]
33%|###2 | 56.0M/170M [00:01<00:02, 43.4MB/s]
38%|###7 | 64.0M/170M [00:01<00:02, 49.0MB/s]
42%|####2 | 72.0M/170M [00:01<00:01, 52.2MB/s]
47%|####7 | 80.0M/170M [00:01<00:01, 53.1MB/s]
52%|#####1 | 88.0M/170M [00:02<00:01, 49.5MB/s]
56%|#####5 | 94.7M/170M [00:02<00:01, 53.8MB/s]
60%|#####9 | 102M/170M [00:02<00:01, 58.6MB/s]
63%|######3 | 108M/170M [00:02<00:01, 52.2MB/s
]
66%|######6 | 113M/170M [00:02<00:01, 50.4MB/s]
71%|####### | 120M/170M [00:02<00:00, 57.6MB/s]
75%|#######5 | 128M/170M [00:02<00:00, 56.6MB/s]
79%|#######9 | 134M/170M [00:02<00:00, 55.1MB/s]
82%|########2 | 140M/170M [00:03<00:00, 52.1MB/s]
85%|########5 | 145M/170M [00:03<00:00, 42.2MB/s]
89%|########9 | 152M/170M [00:03<00:00, 42.5MB/s]
94%|#########4| 160M/170M [00:03<00:00, 44.2MB/s]
99%|#########8| 168M/170M [00:03<00:00, 51.8MB/s]
100%|##########| 170M/170M [00:03<00:00, 48.3MB/s]
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: 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)
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: 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').
@@ -299,7 +299,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 4 minutes 12.708 seconds)
+ **Total running time of the script:** ( 3 minutes 48.442 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 b106ed51b1..4160a7e0cf 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -227,7 +227,7 @@ training. Other models require a full post training calibration.
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MobileNet_V2_Weights.IMAGENET1K_V1`. You can also use `weights=MobileNet_V2_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
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]
47%|####6 | 6.30M/13.6M [00:00<00:00, 47.7MB/s]
80%|######## | 10.9M/13.6M [00:00<00:00, 42.5MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 41.0MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
59%|#####8 | 7.99M/13.6M [00:00<00:00, 55.3MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 80.8MB/s]
@@ -409,7 +409,7 @@ 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)
- 91.1190 91.0954 91.8840 90.7369 0.2098
+ 90.4576 90.4198 93.3698 90.1384 0.3345
@@ -458,7 +458,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 27.939 seconds)
+ **Total running time of the script:** ( 1 minutes 20.934 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 5ca680fea6..0fd39b8617 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
@@ -423,7 +423,7 @@ 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)
- 123.4716 123.4034 127.8491 121.4705 0.8379
+ 121.4166 121.2992 125.6682 120.6915 0.6146
@@ -460,7 +460,7 @@ 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:** ( 2 minutes 39.446 seconds)
+ **Total running time of the script:** ( 2 minutes 35.613 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 21f6554f9a..a88af34d8b 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -257,7 +257,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 58.131 seconds)
+ **Total running time of the script:** ( 2 minutes 4.876 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 734ece0deb..a8d50a3ce2 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
@@ -170,7 +170,7 @@ 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 | 6117/132723 [00:00<00:02, 61159.41KB/s]
11%|# | 14375/132723 [00:00<00:01, 73756.01KB/s]
16%|#6 | 21751/132723 [00:00<00:01, 73587.55KB/s]
23%|##2 | 29905/132723 [00:00<00:01, 76719.15KB/s]
28%|##8 | 37578/132723 [00:00<00:01, 54915.12KB/s]
35%|###4 | 45831/132723 [00:00<00:01, 62348.13KB/s]
40%|###9 | 52744/132723 [00:00<00:01, 60531.76KB/s]
46%|####5 | 60891/132723 [00:00<00:01, 66227.03KB/s]
51%|#####1 | 67919/132723 [00:01<00:01, 63708.72KB/s]
57%|#####7 | 76152/132723 [00:01<00:00, 68816.46KB/s]
63%|######2 | 83294/132723 [00:01<00:00, 58507.35KB/s]
69%|######8 | 91409/132723 [00:01<00:00, 64230.02KB/s]
74%|#######4 | 98288/132723 [00:01<00:00, 51991.29KB/s]
78%|#######8 | 104103/132723 [00:01<00:00, 53071.67KB/s]
85%|########4 | 112282/132723 [00:01<00:00, 60168.35KB/s]
89%|########9
| 118775/132723 [00:01<00:00, 60947.09KB/s]
96%|#########5| 126989/132723 [00:02<00:00, 66670.43KB/s]
100%|##########| 132723/132723 [00:02<00:00, 63348.09KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
5%|5 | 7061/132723 [00:00<00:01, 70603.00KB/s]
12%|#1 | 15712/132723 [00:00<00:01, 79955.80KB/s]
18%|#7 | 23708/132723 [00:00<00:01, 59640.98KB/s]
24%|##4 | 32324/132723 [00:00<00:01, 68318.18KB/s]
31%|### | 40863/132723 [00:00<00:01, 73754.54KB/s]
37%|###7 | 49377/132723 [00:00<00:01, 77314.50KB/s]
44%|####3 | 57952/132723 [00:00<00:00, 79909.66KB/s]
50%|##### | 66587/132723 [00:00<00:00, 81878.30KB/s]
57%|#####6 | 75196/132723 [00:00<00:00, 83156.29KB/s]
63%|######2 | 83615/132723 [00:01<00:00, 77912.83KB/s]
69%|######9 | 92197/132723 [00:01<00:00, 80185.58KB/s]
76%|#######5 | 100789/132723 [00:01<00:00, 81848.19KB/s]
82%|########2 | 109426/132723 [00:01<00:00, 83175.37KB/s]
89%|########8 | 118025/132723 [00:01<00:00, 84005.95KB/s]
95%|#########5| 126667/132723 [00:01<00:00, 84720.83KB/s]
100%|#######
###| 132723/132723 [00:01<00:00, 79461.78KB/s]
@@ -246,7 +246,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 4 minutes 24.382 seconds)
+ **Total running time of the script:** ( 4 minutes 8.435 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 c13a798562..a8f7094d3a 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
@@ -5,28 +5,28 @@
Computation times
=================
-**18:41.806** total execution time for **how_to_deploy_models** files:
+**17:42.479** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 04:24.382 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 04:08.435 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 04:12.708 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:48.442 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:39.446 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:35.613 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:58.131 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 02:04.876 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:27.939 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:20.934 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 01:08.105 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 01:05.981 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``) | 00:59.271 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``) | 00:54.553 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:49.280 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:45.143 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:31.685 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:29.529 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:30.852 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:28.968 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``) | 00:00.007 | 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 172ef0312d..9244615138 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
@@ -463,7 +463,7 @@ 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.zipa3a78f71-d8cf-4767-9bb9-3baef1043d1c from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip4abbc63d-ab71-479f-9533-9b14d243364a 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 a023065b16..e05fbf9d7b 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
=================
-**01:02.783** total execution time for **how_to_extend_tvm** files:
+**00:57.963** 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``) | 00:58.392 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:53.923 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:03.149 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.898 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.233 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.135 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.008 | 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 |
+-------------------------------------------------------------------------------------------------+-----------+--------+
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 4f7e3e3fc2..8cdc17dc4e 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
@@ -220,10 +220,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 24496us [24496us] (48.57%; 48.57%)
- FoldScaleAxis: 25940us [11us] (51.43%; 51.43%)
- FoldConstant: 25930us [1903us] (51.41%; 99.96%)
- InferType: 24027us [24027us] (47.64%; 92.66%)
+ InferType: 23880us [23880us] (49.04%; 49.04%)
+ FoldScaleAxis: 24818us [9us] (50.96%; 50.96%)
+ FoldConstant: 24808us [1785us] (50.94%; 99.96%)
+ InferType: 23024us [23024us] (47.28%; 92.81%)
@@ -262,10 +262,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 24332us [24332us] (48.34%; 48.34%)
- FoldScaleAxis: 26006us [10us] (51.66%; 51.66%)
- FoldConstant: 25996us [2073us] (51.64%; 99.96%)
- InferType: 23923us [23923us] (47.53%; 92.03%)
+ InferType: 23006us [23006us] (48.04%; 48.04%)
+ FoldScaleAxis: 24886us [8us] (51.96%; 51.96%)
+ FoldConstant: 24878us [1785us] (51.95%; 99.97%)
+ InferType: 23093us [23093us] (48.22%; 92.82%)
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 79486389b1..8dca0750ff 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
@@ -331,7 +331,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 53.541183 ms
+ Convolution: 40.979232 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 4602d5cf0d..de9cdf5bff 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
@@ -598,7 +598,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 12.221783 ms
+ conv2d with tensor core: 11.890704 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 ecc88682aa..ec475d3bf7 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -134,8 +134,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.020033
- Baseline: 3.752375
+ Numpy running time: 0.018918
+ Baseline: 3.430278
@@ -227,7 +227,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.348482
+ Opt1: 0.308531
@@ -318,7 +318,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.369061
+ Opt2: 0.342998
@@ -406,7 +406,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.152658
+ Opt3: 0.117401
@@ -523,7 +523,7 @@ flattening.
.. code-block:: none
- Opt4: 0.114943
+ Opt4: 0.108893
@@ -635,7 +635,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.116410
+ Opt5: 0.110965
@@ -748,7 +748,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
.. code-block:: none
- Opt6: 0.152030
+ Opt6: 0.147871
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 7cd5262985..23ebe5a690 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
=================
-**00:38.659** total execution time for **how_to_optimize_operators** files:
+**00:35.911** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:35.436 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.735 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.984 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.995 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.240 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.181 | 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 acc6c504fb..9a0877ad39 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
=================
-**11:15.168** total execution time for **how_to_tune_with_autoscheduler** files:
+**10:59.004** 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``) | 06:57.183 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:51.242 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:53.239 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:47.856 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:18.112 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:14.798 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:35.326 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:35.488 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:15.980 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:15.206 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:15.327 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:14.413 | 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 cf7f0cf5d7..2cfa44e015 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
@@ -209,13 +209,6 @@ file and apply it.
-.. rst-class:: sphx-glr-script-out
-
- .. code-block:: none
-
- .T
-
-
@@ -250,479 +243,577 @@ cooperative fetching, unrolling and operator fusion.
@T.prim_func
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
- blockIdx_x = T.launch_thread("blockIdx.x", 28)
- conv2d_nchw = T.allocate([14], "float32", "local")
- pad_temp_shared = T.allocate([72], "float32", "shared")
- kernel_shared = T.allocate([3072], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 64)
- conv2d_nchw_1 = T.Buffer((14,), data=conv2d_nchw, scope="local", align=32)
+ blockIdx_x = T.launch_thread("blockIdx.x", 16)
+ conv2d_nchw = T.allocate([16], "float32", "local")
+ pad_temp_shared = T.allocate([2592], "float32", "shared")
+ kernel_shared = T.allocate([9216], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 98)
+ conv2d_nchw_1 = T.Buffer((64,), data=conv2d_nchw, scope="local", align=32)
conv2d_nchw_1[0] = T.float32(0)
+ conv2d_nchw_1[8] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
+ conv2d_nchw_1[9] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
+ conv2d_nchw_1[10] = T.float32(0)
conv2d_nchw_1[3] = T.float32(0)
+ conv2d_nchw_1[11] = T.float32(0)
conv2d_nchw_1[4] = T.float32(0)
+ conv2d_nchw_1[12] = T.float32(0)
conv2d_nchw_1[5] = T.float32(0)
+ conv2d_nchw_1[13] = T.float32(0)
conv2d_nchw_1[6] = T.float32(0)
+ conv2d_nchw_1[14] = T.float32(0)
conv2d_nchw_1[7] = T.float32(0)
- conv2d_nchw_1[8] = T.float32(0)
- conv2d_nchw_1[9] = T.float32(0)
- conv2d_nchw_1[10] = T.float32(0)
- conv2d_nchw_1[11] = T.float32(0)
- conv2d_nchw_1[12] = T.float32(0)
- conv2d_nchw_1[13] = T.float32(0)
- for rc_outer_outer, ry_outer_outer in T.grid(64, 3):
- cse_var_2: T.int32 = rc_outer_outer * 72
- cse_var_1: T.int32 = ry_outer_outer * 3
- pad_temp_shared_1 = T.Buffer((72,), data=pad_temp_shared, scope="shared")
- with T.launch_thread("threadIdx.x", 64) as threadIdx_x_1:
- data_1 = T.Buffer((25088,), data=data.data)
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[rc_outer_outer * 392 + threadIdx_x_1 * 4 // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + threadIdx_x_1 * 4 % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 1] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 1) % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 2] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 2) % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 3] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 3) % 9 - 8], T.float32(0))
+ conv2d_nchw_1[15] = T.float32(0)
+ for rc_outer_outer in range(16):
+ cse_var_2: T.int32 = rc_outer_outer * 1568
+ cse_var_1: T.int32 = rc_outer_outer * 288
threadIdx_x_1 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((3072,), data=kernel_shared, scope="shared")
+ pad_temp_shared_1 = T.Buffer((2592,), data=pad_temp_shared, scope="shared")
+ data_1 = T.Buffer((25088,), data=data.data)
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(9 <= threadIdx_x_1 % 81 and threadIdx_x_1 % 81 < 72 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 // 81 * 49 + threadIdx_x_1 % 81 // 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 98] = T.if_then_else(9 <= (threadIdx_x_1 + 17) % 81 and (threadIdx_x_1 + 17) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 98) // 81 * 49 + (threadIdx_x_1 + 17) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 196] = T.if_then_else(9 <= (threadIdx_x_1 + 34) % 81 and (threadIdx_x_1 + 34) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 196) // 81 * 49 + (threadIdx_x_1 + 34) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 294] = T.if_then_else(9 <= (threadIdx_x_1 + 51) % 81 and (threadIdx_x_1 + 51) % 81 < 72 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 294) // 81 * 49 + (threadIdx_x_1 + 51) % 81 // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 392] = T.if_then_else(9 <= (threadIdx_x_1 + 68) % 81 and (threadIdx_x_1 + 68) % 81 < 72 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 392) // 81 * 49 + (threadIdx_x_1 + 68) % 81 // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 490] = T.if_then_else(9 <= (threadIdx_x_1 + 4) % 81 and (threadIdx_x_1 + 4) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 490) // 81 * 49 + (threadIdx_x_1 + 4) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 588] = T.if_then_else(9 <= (threadIdx_x_1 + 21) % 81 and (threadIdx_x_1 + 21) % 81 < 72 and 1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 588) // 81 * 49 + (threadIdx_x_1 + 21) % 81 // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 686] = T.if_then_else(9 <= (threadIdx_x_1 + 38) % 81 and (threadIdx_x_1 + 38) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 686) // 81 * 49 + (threadIdx_x_1 + 38) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 784] = T.if_then_else(9 <= (threadIdx_x_1 + 55) % 81 and (threadIdx_x_1 + 55) % 81 < 72 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 784) // 81 * 49 + (threadIdx_x_1 + 55) % 81 // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 882] = T.if_then_else(1 <= (threadIdx_x_1 // 9 + 8) % 9 and (threadIdx_x_1 + 72) % 81 < 72 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 882) // 81 * 49 + (threadIdx_x_1 // 9 + 8) % 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 980] = T.if_then_else(9 <= (threadIdx_x_1 + 8) % 81 and (threadIdx_x_1 + 8) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 980) // 81 * 49 + (threadIdx_x_1 + 8) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1078] = T.if_then_else(9 <= (threadIdx_x_1 + 25) % 81 and (threadIdx_x_1 + 25) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1078) // 81 * 49 + (threadIdx_x_1 + 25) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1176] = T.if_then_else(9 <= (threadIdx_x_1 + 42) % 81 and (threadIdx_x_1 + 42) % 81 < 72 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1176) // 81 * 49 + (threadIdx_x_1 + 42) % 81 // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1274] = T.if_then_else(9 <= (threadIdx_x_1 + 59) % 81 and (threadIdx_x_1 + 59) % 81 < 72 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1274) // 81 * 49 + (threadIdx_x_1 + 59) % 81 // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1372] = T.if_then_else(9 <= (threadIdx_x_1 + 76) % 81 and (threadIdx_x_1 + 76) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1372) // 81 * 49 + (threadIdx_x_1 + 76) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1470] = T.if_then_else(9 <= (threadIdx_x_1 + 12) % 81 and (threadIdx_x_1 + 12) % 81 < 72 and 1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1470) // 81 * 49 + (threadIdx_x_1 + 12) % 81 // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1568] = T.if_then_else(9 <= (threadIdx_x_1 + 29) % 81 and (threadIdx_x_1 + 29) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1568) // 81 * 49 + (threadIdx_x_1 + 29) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1666] = T.if_then_else(9 <= (threadIdx_x_1 + 46) % 81 and (threadIdx_x_1 + 46) % 81 < 72 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1666) // 81 * 49 + (threadIdx_x_1 + 46) % 81 // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1764] = T.if_then_else(1 <= (threadIdx_x_1 // 9 + 7) % 9 and (threadIdx_x_1 + 63) % 81 < 72 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1764) // 81 * 49 + (threadIdx_x_1 // 9 + 7) % 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1862] = T.if_then_else(9 <= (threadIdx_x_1 + 80) % 81 and (threadIdx_x_1 + 80) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1862) // 81 * 49 + (threadIdx_x_1 + 80) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1960] = T.if_then_else(9 <= (threadIdx_x_1 + 16) % 81 and (threadIdx_x_1 + 16) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1960) // 81 * 49 + (threadIdx_x_1 + 16) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2058] = T.if_then_else(9 <= (threadIdx_x_1 + 33) % 81 and (threadIdx_x_1 + 33) % 81 < 72 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2058) // 81 * 49 + (threadIdx_x_1 + 33) % 81 // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2156] = T.if_then_else(9 <= (threadIdx_x_1 + 50) % 81 and (threadIdx_x_1 + 50) % 81 < 72 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2156) // 81 * 49 + (threadIdx_x_1 + 50) % 81 // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2254] = T.if_then_else(9 <= (threadIdx_x_1 + 67) % 81 and (threadIdx_x_1 + 67) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2254) // 81 * 49 + (threadIdx_x_1 + 67) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2352] = T.if_then_else(9 <= (threadIdx_x_1 + 3) % 81 and (threadIdx_x_1 + 3) % 81 < 72 and 1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2352) // 81 * 49 + (threadIdx_x_1 + 3) % 81 // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2450] = T.if_then_else(9 <= (threadIdx_x_1 + 20) % 81 and (threadIdx_x_1 + 20) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2450) // 81 * 49 + (threadIdx_x_1 + 20) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ if T.likely(threadIdx_x_1 < 44):
+ pad_temp_shared_1[threadIdx_x_1 + 2548] = T.if_then_else(threadIdx_x_1 < 35 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2548) // 81 * 49 + (threadIdx_x_1 + 37) // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
+ threadIdx_x_2 = T.env_thread("threadIdx.x")
+ kernel_shared_1 = T.Buffer((9216,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 64) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 64) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 128) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 128) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 192] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 36864]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 256) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 256) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 320) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 320) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 384] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 73728]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 448) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 448) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 512) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 512) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 576] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 110592]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 640) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 640) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 704) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 704) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 768] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 147456]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 832) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 832) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 896) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 896) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 960] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 184320]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1024) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1024) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1088) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1088) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1152] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 221184]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1216) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1216) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1280) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1280) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1344] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 258048]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1408) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1408) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1472) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1472) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1536] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 294912]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1600) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1600) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1664) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1664) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1728] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 331776]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1792) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1792) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1856) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1856) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1920] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 368640]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1984) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1984) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2048) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2048) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2112] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 405504]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2176) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2176) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2240) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2240) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2304] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 442368]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2368) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2368) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2432) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2432) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2496] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 479232]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2560) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2560) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2624) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2624) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2688] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 516096]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2752) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2752) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2816) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2816) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2880] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 552960]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2944) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2944) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 3008) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 3008) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 47]
- for i1_inner, i3_inner in T.grid(2, 7):
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x * 147456 + cse_var_1 + threadIdx_x_2]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 98] = kernel_1[blockIdx_x * 147456 + cse_var_1 + threadIdx_x_2 + 98]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 196] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 196) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 196) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 294] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 294) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 6]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 392] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 104]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 490] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 490) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 202) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 588] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 588) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 12]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 686] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 686) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 110]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 784] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 784) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 208) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 882] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 882) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 18]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 980] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 980) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 116]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1078] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1078) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 214) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1176] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1176) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 24]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1274] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1274) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 122]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1372] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1372) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 220) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1470] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1470) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 30]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1568] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1568) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 128]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1666] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1666) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 226) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1764] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1764) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 36]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1862] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1862) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 134]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1960] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1960) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 232) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2058] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2058) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 42]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2156] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2156) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 140]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2254] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2254) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 238) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2352] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2352) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 48]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2450] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2450) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 146]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2548] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2548) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 244) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2646] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2646) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 54]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2744] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2744) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 152]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2842] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2842) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 250) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2940] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2940) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 60]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3038] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3038) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 158]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3136] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3136) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 256) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3234] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3234) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 66]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3332] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3332) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 164]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3430] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3430) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 262) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3528] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3528) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 72]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3626] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3626) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 170]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3724] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3724) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 268) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3822] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3822) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 78]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3920] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3920) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 176]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4018] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4018) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 274) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4116] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4116) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 84]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4214] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4214) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 182]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4312] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4312) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 280) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4410] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4410) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 90]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4508] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4508) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 188]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4606] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4606) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 286) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4704] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4704) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 96]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4802] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4802) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 194) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4900] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4900) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 4]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4998] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4998) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 102]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5096] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5096) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 200) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5194] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5194) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 10]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5292] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5292) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 108]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5390] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5390) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 206) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5488] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5488) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 16]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5586] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5586) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 114]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5684] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5684) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 212) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5782] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5782) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 22]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5880] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5880) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 120]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5978] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5978) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 218) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6076] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6076) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 28]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6174] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6174) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 126]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6272] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6272) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 224) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6370] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6370) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 34]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6468] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6468) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 132]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6566] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6566) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 230) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6664] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6664) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 40]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6762] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6762) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 138]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6860] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6860) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 236) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6958] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6958) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 46]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7056] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7056) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 144]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7154] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7154) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 242) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7252] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7252) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 52]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7350] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7350) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 150]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7448] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7448) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 248) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7546] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7546) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 58]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7644] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7644) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 156]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7742] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7742) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 254) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7840] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7840) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 64]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7938] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7938) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 162]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8036] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8036) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 260) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8134] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8134) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 70]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8232] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8232) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 168]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8330] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8330) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 266) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8428] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8428) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 76]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8526] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8526) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 174]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8624] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8624) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 272) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8722] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8722) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 82]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8820] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8820) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 180]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8918] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8918) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 278) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 9016] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 9016) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 88]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 9114] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 9114) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 186]
+ with T.launch_thread(threadIdx_x_2, 98):
+ if T.likely(threadIdx_x_2 < 4):
+ kernel_shared_1[threadIdx_x_2 + 9212] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 9212) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 284]
+ for rc_outer_inner in range(16):
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4608]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 9]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4617]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 288]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4896]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 297]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4905]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 576]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5184]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 585]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5193]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 864]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5472]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 873]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5481]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1152]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5760]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1161]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5769]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1440]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6048]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1449]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6057]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1728]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6336]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1737]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6345]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2016]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6624]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2025]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6633]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4609]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 10]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4618]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 289]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4897]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 298]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4906]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 577]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5185]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 586]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5194]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 865]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5473]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 874]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5482]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1153]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5761]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1162]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5770]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1441]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6049]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1450]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6058]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1729]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6337]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1738]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6346]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2017]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6625]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2026]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6634]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4610]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 11]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4619]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 290]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4898]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 299]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4907]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 578]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5186]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 587]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5195]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 866]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5474]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 875]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5483]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1154]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5762]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1163]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5771]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1442]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6050]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1451]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6059]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1730]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6338]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1739]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6347]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2018]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6626]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2027]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6635]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 3]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4611]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 12]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4620]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 291]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4899]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 300]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4908]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 579]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5187]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 588]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5196]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 867]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5475]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 876]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5484]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1155]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5763]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1164]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5772]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1443]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6051]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1452]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6060]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1731]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6339]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1740]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6348]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2019]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6627]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2028]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6636]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4612]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 13]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4621]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 292]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4900]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 301]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4909]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 580]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5188]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 589]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5197]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 868]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5476]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 877]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5485]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1156]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5764]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1165]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5773]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1444]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6052]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1453]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6061]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1732]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6340]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1741]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6349]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2020]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6628]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2029]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6637]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4613]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 14]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4622]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 293]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4901]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 302]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4910]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 581]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5189]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 590]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5198]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 869]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5477]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 878]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5486]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1157]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5765]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1166]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5774]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1445]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6053]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1454]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6062]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1733]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6341]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1742]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6350]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2021]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6629]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2030]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6638]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4614]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 15]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4623]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 294]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4902]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 303]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4911]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 582]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5190]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 591]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5199]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 870]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5478]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 879]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5487]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1158]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5766]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1167]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5775]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1446]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6054]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1455]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6063]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1734]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6342]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1743]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6351]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2022]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6630]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2031]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6639]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 7]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4615]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 16]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4624]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 295]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4903]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 304]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4912]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 583]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5191]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 592]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5200]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 871]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5479]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 880]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5488]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1159]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5767]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1168]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5776]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1447]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6055]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1456]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6064]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1735]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6343]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1744]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6352]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2023]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6631]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2032]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6640]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 8]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4616]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 17]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4625]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 296]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4904]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 305]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4913]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 584]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5192]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 593]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5201]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 872]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5480]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 881]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5489]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1160]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5768]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1169]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5777]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1448]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6056]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1457]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6065]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1736]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6344]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1745]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6353]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2024]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6632]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2033]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6641]
+ for i1_inner in range(8):
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x // 7 * 6272 + threadIdx_x * 98 + i1_inner * 49 + blockIdx_x % 7 * 7 + i3_inner] = T.max(conv2d_nchw_1[i1_inner * 7 + i3_inner] + bias_1[blockIdx_x // 7 * 128 + threadIdx_x * 2 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 1568 + threadIdx_x // 49 * 392 + i1_inner * 49 + threadIdx_x % 49] = T.max(conv2d_nchw_1[i1_inner] + bias_1[blockIdx_x * 32 + threadIdx_x // 49 * 8 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 1568 + threadIdx_x // 49 * 392 + i1_inner * 49 + threadIdx_x % 49 + 784] = T.max(conv2d_nchw_1[i1_inner + 8] + bias_1[blockIdx_x * 32 + threadIdx_x // 49 * 8 + i1_inner + 16], T.float32(0))
@@ -772,7 +863,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.339 ms
+ Execution time of this operator: 0.268 ms
@@ -821,35 +912,35 @@ 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=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=64)
- 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=8)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=2)
+ conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=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=1)
+ conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
- conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
- conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, 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=2)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=16)
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=1)
+ conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
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=2)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
- compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
+ compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=8)
+ 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=2)
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=1)
+ compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
- compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
- compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_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)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -869,14 +960,14 @@ 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=64)
+ kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=4)
+ 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=64)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
- s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+ 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)
CUDA source code:
@@ -901,430 +992,449 @@ 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__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[14];
- __shared__ float pad_temp_shared[72];
- __shared__ float kernel_shared[3072];
+ extern "C" __global__ void __launch_bounds__(98) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[16];
+ __shared__ float pad_temp_shared[2592];
+ __shared__ float kernel_shared[9216];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
conv2d_nchw[7] = 0.000000e+00f;
- conv2d_nchw[8] = 0.000000e+00f;
- conv2d_nchw[9] = 0.000000e+00f;
- conv2d_nchw[10] = 0.000000e+00f;
- conv2d_nchw[11] = 0.000000e+00f;
- conv2d_nchw[12] = 0.000000e+00f;
- conv2d_nchw[13] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
- for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
- __syncthreads();
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[(((int)threadIdx.x) * 4)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 4) % 9))) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) * 4) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 1)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 1) % 9))) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 1) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 2)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 2) % 9))) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 2) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 3)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 3) % 9))) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 3) % 9)) - 8)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 64) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 128) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((((((int)threadIdx.x) + 256) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 320) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((((((int)threadIdx.x) + 448) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 512) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((((((int)threadIdx.x) + 640) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 704) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
- kernel_shared[(((((((int)threadIdx.x) + 832) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 896) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 896) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
- kernel_shared[(((((((int)threadIdx.x) + 1024) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1088) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
- kernel_shared[(((((((int)threadIdx.x) + 1216) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1280) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
- kernel_shared[(((((((int)threadIdx.x) + 1408) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1472) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 294912)];
- kernel_shared[(((((((int)threadIdx.x) + 1600) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1664) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 331776)];
- kernel_shared[(((((((int)threadIdx.x) + 1792) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1856) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 368640)];
- kernel_shared[(((((((int)threadIdx.x) + 1984) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2048) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 405504)];
- kernel_shared[(((((((int)threadIdx.x) + 2176) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2240) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 442368)];
- kernel_shared[(((((((int)threadIdx.x) + 2368) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2432) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 479232)];
- kernel_shared[(((((((int)threadIdx.x) + 2560) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2624) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 516096)];
- kernel_shared[(((((((int)threadIdx.x) + 2752) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2816) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 552960)];
- kernel_shared[(((((((int)threadIdx.x) + 2944) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 3008) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+ conv2d_nchw[15] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[((int)threadIdx.x)] = (((((9 <= (((int)threadIdx.x) % 81)) && ((((int)threadIdx.x) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 81) * 49)) + (((((int)threadIdx.x) % 81) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 98)] = (((((9 <= ((((int)threadIdx.x) + 17) % 81)) && (((((int)threadIdx.x) + 17) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 98) / 81) * 49)) + ((((((int)threadIdx.x) + 17) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((9 <= ((((int)threadIdx.x) + 34) % 81)) && (((((int)threadIdx.x) + 34) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 196) / 81) * 49)) + ((((((int)threadIdx.x) + 34) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 294)] = (((((9 <= ((((int)threadIdx.x) + 51) % 81)) && (((((int)threadIdx.x) + 51) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 294) / 81) * 49)) + ((((((int)threadIdx.x) + 51) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((9 <= ((((int)threadIdx.x) + 68) % 81)) && (((((int)threadIdx.x) + 68) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 392) / 81) * 49)) + ((((((int)threadIdx.x) + 68) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 490)] = (((((9 <= ((((int)threadIdx.x) + 4) % 81)) && (((((int)threadIdx.x) + 4) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 490) / 81) * 49)) + ((((((int)threadIdx.x) + 4) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 588)] = (((((9 <= ((((int)threadIdx.x) + 21) % 81)) && (((((int)threadIdx.x) + 21) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 588) / 81) * 49)) + ((((((int)threadIdx.x) + 21) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 686)] = (((((9 <= ((((int)threadIdx.x) + 38) % 81)) && (((((int)threadIdx.x) + 38) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 686) / 81) * 49)) + ((((((int)threadIdx.x) + 38) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((9 <= ((((int)threadIdx.x) + 55) % 81)) && (((((int)threadIdx.x) + 55) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 784) / 81) * 49)) + ((((((int)threadIdx.x) + 55) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 882)] = (((((1 <= (((((int)threadIdx.x) / 9) + 8) % 9)) && (((((int)threadIdx.x) + 72) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 882) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 8) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 980)] = (((((9 <= ((((int)threadIdx.x) + 8) % 81)) && (((((int)threadIdx.x) + 8) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 980) / 81) * 49)) + ((((((int)threadIdx.x) + 8) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1078)] = (((((9 <= ((((int)threadIdx.x) + 25) % 81)) && (((((int)threadIdx.x) + 25) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1078) / 81) * 49)) + ((((((int)threadIdx.x) + 25) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((9 <= ((((int)threadIdx.x) + 42) % 81)) && (((((int)threadIdx.x) + 42) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1176) / 81) * 49)) + ((((((int)threadIdx.x) + 42) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1274)] = (((((9 <= ((((int)threadIdx.x) + 59) % 81)) && (((((int)threadIdx.x) + 59) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1274) / 81) * 49)) + ((((((int)threadIdx.x) + 59) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1372)] = (((((9 <= ((((int)threadIdx.x) + 76) % 81)) && (((((int)threadIdx.x) + 76) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1372) / 81) * 49)) + ((((((int)threadIdx.x) + 76) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1470)] = (((((9 <= ((((int)threadIdx.x) + 12) % 81)) && (((((int)threadIdx.x) + 12) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1470) / 81) * 49)) + ((((((int)threadIdx.x) + 12) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1568)] = (((((9 <= ((((int)threadIdx.x) + 29) % 81)) && (((((int)threadIdx.x) + 29) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1568) / 81) * 49)) + ((((((int)threadIdx.x) + 29) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1666)] = (((((9 <= ((((int)threadIdx.x) + 46) % 81)) && (((((int)threadIdx.x) + 46) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1666) / 81) * 49)) + ((((((int)threadIdx.x) + 46) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1764)] = (((((1 <= (((((int)threadIdx.x) / 9) + 7) % 9)) && (((((int)threadIdx.x) + 63) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1764) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 7) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1862)] = (((((9 <= ((((int)threadIdx.x) + 80) % 81)) && (((((int)threadIdx.x) + 80) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1862) / 81) * 49)) + ((((((int)threadIdx.x) + 80) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1960)] = (((((9 <= ((((int)threadIdx.x) + 16) % 81)) && (((((int)threadIdx.x) + 16) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1960) / 81) * 49)) + ((((((int)threadIdx.x) + 16) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2058)] = (((((9 <= ((((int)threadIdx.x) + 33) % 81)) && (((((int)threadIdx.x) + 33) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2058) / 81) * 49)) + ((((((int)threadIdx.x) + 33) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2156)] = (((((9 <= ((((int)threadIdx.x) + 50) % 81)) && (((((int)threadIdx.x) + 50) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2156) / 81) * 49)) + ((((((int)threadIdx.x) + 50) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2254)] = (((((9 <= ((((int)threadIdx.x) + 67) % 81)) && (((((int)threadIdx.x) + 67) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2254) / 81) * 49)) + ((((((int)threadIdx.x) + 67) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2352)] = (((((9 <= ((((int)threadIdx.x) + 3) % 81)) && (((((int)threadIdx.x) + 3) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2352) / 81) * 49)) + ((((((int)threadIdx.x) + 3) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2450)] = (((((9 <= ((((int)threadIdx.x) + 20) % 81)) && (((((int)threadIdx.x) + 20) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2450) / 81) * 49)) + ((((((int)threadIdx.x) + 20) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 44) {
+ pad_temp_shared[(((int)threadIdx.x) + 2548)] = ((((((int)threadIdx.x) < 35) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2548) / 81) * 49)) + (((((int)threadIdx.x) + 37) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
}
- }
- for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 147456) + (rc_outer_outer * 288)) + ((int)threadIdx.x))];
+ kernel_shared[(((int)threadIdx.x) + 98)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 98)];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 196) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 196) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 294)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 294) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 6)];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 104)];
+ kernel_shared[(((int)threadIdx.x) + 490)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 490) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 202) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 588)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 588) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 12)];
+ kernel_shared[(((int)threadIdx.x) + 686)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 686) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 110)];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 208) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 882)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 882) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 18)];
+ kernel_shared[(((int)threadIdx.x) + 980)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 980) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 116)];
+ kernel_shared[(((int)threadIdx.x) + 1078)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1078) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 214) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1176) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 1274)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1274) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 122)];
+ kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1372) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 220) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1470)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1470) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 30)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1568) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 128)];
+ kernel_shared[(((int)threadIdx.x) + 1666)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1666) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 226) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1764)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1764) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 36)];
+ kernel_shared[(((int)threadIdx.x) + 1862)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1862) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 134)];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1960) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 232) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2058)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2058) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 42)];
+ kernel_shared[(((int)threadIdx.x) + 2156)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2156) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 140)];
+ kernel_shared[(((int)threadIdx.x) + 2254)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2254) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 238) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2352) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 2450)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2450) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 146)];
+ kernel_shared[(((int)threadIdx.x) + 2548)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2548) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 244) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2646)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2646) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 54)];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2744) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 152)];
+ kernel_shared[(((int)threadIdx.x) + 2842)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2842) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 250) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2940)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2940) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 60)];
+ kernel_shared[(((int)threadIdx.x) + 3038)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3038) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 158)];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3136) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 256) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3234)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3234) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 66)];
+ kernel_shared[(((int)threadIdx.x) + 3332)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3332) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 164)];
+ kernel_shared[(((int)threadIdx.x) + 3430)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3430) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 262) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3528)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3528) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 3626)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3626) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 170)];
+ kernel_shared[(((int)threadIdx.x) + 3724)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3724) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 268) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3822)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3822) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 78)];
+ kernel_shared[(((int)threadIdx.x) + 3920)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3920) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 176)];
+ kernel_shared[(((int)threadIdx.x) + 4018)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4018) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 274) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4116)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4116) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 84)];
+ kernel_shared[(((int)threadIdx.x) + 4214)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4214) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 182)];
+ kernel_shared[(((int)threadIdx.x) + 4312)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4312) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 280) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4410)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4410) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 90)];
+ kernel_shared[(((int)threadIdx.x) + 4508)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4508) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 188)];
+ kernel_shared[(((int)threadIdx.x) + 4606)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4606) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 286) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4704)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4704) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 96)];
+ kernel_shared[(((int)threadIdx.x) + 4802)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4802) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 194) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4900)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4900) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 4)];
+ kernel_shared[(((int)threadIdx.x) + 4998)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4998) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 102)];
+ kernel_shared[(((int)threadIdx.x) + 5096)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5096) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 200) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 5194)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5194) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 10)];
+ kernel_shared[(((int)threadIdx.x) + 5292)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5292) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 108)];
+ kernel_shared[(((int)threadIdx.x) + 5390)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5390) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 206) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 5488)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5488) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 16)];
+ kernel_shared[(((int)threadIdx.x) + 5586)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5586) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 114)];
+ kernel_shared[(((int)threadIdx.x) + 5684)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5684) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 212) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 5782)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5782) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 22)];
+ kernel_shared[(((int)threadIdx.x) + 5880)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5880) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 120)];
+ kernel_shared[(((int)threadIdx.x) + 5978)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5978) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 218) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 6076)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6076) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 28)];
+ kernel_shared[(((int)threadIdx.x) + 6174)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6174) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 126)];
+ kernel_shared[(((int)threadIdx.x) + 6272)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6272) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 224) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 6370)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6370) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 34)];
+ kernel_shared[(((int)threadIdx.x) + 6468)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6468) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 132)];
+ kernel_shared[(((int)threadIdx.x) + 6566)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6566) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 230) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 6664)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6664) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 40)];
+ kernel_shared[(((int)threadIdx.x) + 6762)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6762) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 138)];
+ kernel_shared[(((int)threadIdx.x) + 6860)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6860) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 236) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 6958)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6958) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 46)];
+ kernel_shared[(((int)threadIdx.x) + 7056)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7056) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 7154)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7154) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 242) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 7252)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7252) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 52)];
+ kernel_shared[(((int)threadIdx.x) + 7350)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7350) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 150)];
+ kernel_shared[(((int)threadIdx.x) + 7448)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7448) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 248) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 7546)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7546) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 58)];
+ kernel_shared[(((int)threadIdx.x) + 7644)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7644) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 156)];
+ kernel_shared[(((int)threadIdx.x) + 7742)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7742) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 254) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 7840)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7840) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 64)];
+ kernel_shared[(((int)threadIdx.x) + 7938)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7938) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 162)];
+ kernel_shared[(((int)threadIdx.x) + 8036)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8036) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 260) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 8134)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8134) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 70)];
+ kernel_shared[(((int)threadIdx.x) + 8232)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8232) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 168)];
+ kernel_shared[(((int)threadIdx.x) + 8330)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8330) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 266) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 8428)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8428) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 76)];
+ kernel_shared[(((int)threadIdx.x) + 8526)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8526) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 174)];
+ kernel_shared[(((int)threadIdx.x) + 8624)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8624) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 272) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 8722)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8722) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 82)];
+ kernel_shared[(((int)threadIdx.x) + 8820)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8820) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 180)];
+ kernel_shared[(((int)threadIdx.x) + 8918)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8918) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 278) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 9016)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 9016) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 88)];
+ kernel_shared[(((int)threadIdx.x) + 9114)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 9114) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 186)];
+ if (((int)threadIdx.x) < 4) {
+ kernel_shared[(((int)threadIdx.x) + 9212)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 9212) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 284)];
}
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 16; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18))]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4608)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 9)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4617)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 288)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4896)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 297)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4905)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 576)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5184)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 585)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5193)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 864)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5472)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 873)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5481)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1152)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5760)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1161)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5769)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1440)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6048)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1449)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6057)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1728)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6336)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1737)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6345)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2016)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6624)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2025)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6633)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4609)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 10)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4618)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 289)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4897)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 298)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4906)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 577)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5185)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 586)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5194)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 865)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5473)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 874)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5482)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1153)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5761)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1162)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5770)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1441)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6049)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1450)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6058)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1729)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6337)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1738)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6346)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2017)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6625)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2026)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6634)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4610)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 11)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4619)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 290)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4898)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 299)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4907)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 578)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5186)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 587)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5195)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 866)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5474)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 875)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5483)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1154)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5762)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1163)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5771)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1442)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6050)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1451)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6059)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1730)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6338)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1739)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6347)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2018)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6626)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2027)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6635)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 3)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4611)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 12)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4620)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 291)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4899)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 300)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4908)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 579)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5187)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 588)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5196)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 867)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5475)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 876)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5484)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1155)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5763)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1164)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5772)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1443)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6051)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1452)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6060)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1731)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6339)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1740)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6348)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2019)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6627)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2028)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6636)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4612)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 13)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4621)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 292)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4900)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 301)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4909)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 580)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5188)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 589)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5197)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 868)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5476)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 877)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5485)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1156)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5764)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1165)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5773)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1444)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6052)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1453)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6061)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1732)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6340)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1741)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6349)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2020)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6628)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2029)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6637)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4613)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 14)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4622)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 293)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4901)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 302)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4910)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 581)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5189)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 590)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5198)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 869)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5477)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 878)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5486)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1157)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5765)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1166)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5774)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1445)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6053)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1454)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6062)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1733)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6341)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1742)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6350)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2021)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6629)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2030)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6638)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4614)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 15)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4623)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 294)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4902)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 303)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4911)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 582)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5190)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 591)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5199)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 870)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5478)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 879)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5487)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1158)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5766)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1167)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5775)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1446)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6054)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1455)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6063)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1734)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6342)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1743)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6351)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2022)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6630)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2031)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6639)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 7)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4615)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 16)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4624)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 295)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4903)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 304)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4912)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 583)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5191)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 592)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5200)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 871)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5479)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 880)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5488)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1159)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5767)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1168)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5776)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1447)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6055)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1456)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6064)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1735)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6343)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1744)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6352)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2023)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6631)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2032)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6640)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 8)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4616)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 17)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4625)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 296)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4904)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 305)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4913)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 584)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5192)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 593)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5201)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 872)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5480)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 881)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5489)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1160)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5768)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1169)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5777)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1448)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6056)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1457)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6065)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1736)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6344)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1745)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6353)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2024)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6632)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2033)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6641)]));
+ }
+ }
+ for (int i1_inner = 0; i1_inner < 8; ++i1_inner) {
+ compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 392)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 8)) + i1_inner)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 392)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 784)] = max((conv2d_nchw[(i1_inner + 8)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 8)) + i1_inner) + 16)]), 0.000000e+00f);
}
}
@@ -1376,7 +1486,7 @@ In the example below we resume the status and do more 5 trials.
Resume search:
Get devices for measurement successfully!
-
+ .T
@@ -1384,7 +1494,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 6 minutes 57.183 seconds)
+ **Total running time of the script:** ( 6 minutes 51.242 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 e97a076f55..56621c298c 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)
- 8.1218 8.1208 8.1314 8.1133 0.0074
+ 8.1176 8.1192 8.1229 8.1106 0.0052
@@ -675,7 +675,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 18.112 seconds)
+ **Total running time of the script:** ( 1 minutes 14.798 seconds)
.. _sphx_glr_download_how_to_tune_with_autoscheduler_tune_network_cuda.py:
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 6ebf9c8dc3..c012336173 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)
- 798.0223 799.0309 802.9418 792.0942 4.4856
+ 758.5152 758.9505 759.4899 757.1053 1.0210
@@ -694,7 +694,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 53.239 seconds)
+ **Total running time of the script:** ( 1 minutes 47.856 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 8728935018..23303ecd33 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
@@ -389,26 +389,407 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
@T.prim_func
def main(placeholder: T.Buffer((128, 256), "float32"), placeholder_1: T.Buffer((4916, 16, 1), "float32"), placeholder_2: T.Buffer((4916,), "int32"), placeholder_3: T.Buffer((33,), "int32"), placeholder_4: T.Buffer((128, 512), "float32"), compute: T.Buffer((128, 512), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
- for i0_outer_i1_outer_fused in T.parallel(64):
- compute_1 = T.allocate([1024], "float32", "global")
- compute_2 = T.Buffer((1024,), data=compute_1)
- for nb_j_inner in range(2):
- for i_inner_init, j_init in T.grid(32, 16):
- compute_2[i_inner_init * 32 + nb_j_inner * 16 + j_init] = T.float32(0)
- for elem_idx, i_inner, j in T.grid(T.Let(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1], where={cse_var_1: i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner}), 32, 16):
- cse_var_1 = T.int32()
+ for i0_outer_i1_outer_fused in T.parallel(128):
+ compute_1 = T.allocate([512], "float32", "global")
+ compute_2 = T.Buffer((512,), data=compute_1)
+ for i_outer_inner, nb_j_inner in T.grid(2, 2):
+ cse_var_2: T.int32 = i_outer_inner * 256 + nb_j_inner * 16
+ cse_var_1: T.int32 = i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner
+ compute_2[cse_var_2] = T.float32(0)
+ compute_2[cse_var_2 + 1] = T.float32(0)
+ compute_2[cse_var_2 + 2] = T.float32(0)
+ compute_2[cse_var_2 + 3] = T.float32(0)
+ compute_2[cse_var_2 + 4] = T.float32(0)
+ compute_2[cse_var_2 + 5] = T.float32(0)
+ compute_2[cse_var_2 + 6] = T.float32(0)
+ compute_2[cse_var_2 + 7] = T.float32(0)
+ compute_2[cse_var_2 + 8] = T.float32(0)
+ compute_2[cse_var_2 + 9] = T.float32(0)
+ compute_2[cse_var_2 + 10] = T.float32(0)
+ compute_2[cse_var_2 + 11] = T.float32(0)
+ compute_2[cse_var_2 + 12] = T.float32(0)
+ compute_2[cse_var_2 + 13] = T.float32(0)
+ compute_2[cse_var_2 + 14] = T.float32(0)
+ compute_2[cse_var_2 + 15] = T.float32(0)
+ compute_2[cse_var_2 + 32] = T.float32(0)
+ compute_2[cse_var_2 + 33] = T.float32(0)
+ compute_2[cse_var_2 + 34] = T.float32(0)
+ compute_2[cse_var_2 + 35] = T.float32(0)
+ compute_2[cse_var_2 + 36] = T.float32(0)
+ compute_2[cse_var_2 + 37] = T.float32(0)
+ compute_2[cse_var_2 + 38] = T.float32(0)
+ compute_2[cse_var_2 + 39] = T.float32(0)
+ compute_2[cse_var_2 + 40] = T.float32(0)
+ compute_2[cse_var_2 + 41] = T.float32(0)
+ compute_2[cse_var_2 + 42] = T.float32(0)
+ compute_2[cse_var_2 + 43] = T.float32(0)
+ compute_2[cse_var_2 + 44] = T.float32(0)
+ compute_2[cse_var_2 + 45] = T.float32(0)
+ compute_2[cse_var_2 + 46] = T.float32(0)
+ compute_2[cse_var_2 + 47] = T.float32(0)
+ compute_2[cse_var_2 + 64] = T.float32(0)
+ compute_2[cse_var_2 + 65] = T.float32(0)
+ compute_2[cse_var_2 + 66] = T.float32(0)
+ compute_2[cse_var_2 + 67] = T.float32(0)
+ compute_2[cse_var_2 + 68] = T.float32(0)
+ compute_2[cse_var_2 + 69] = T.float32(0)
+ compute_2[cse_var_2 + 70] = T.float32(0)
+ compute_2[cse_var_2 + 71] = T.float32(0)
+ compute_2[cse_var_2 + 72] = T.float32(0)
+ compute_2[cse_var_2 + 73] = T.float32(0)
+ compute_2[cse_var_2 + 74] = T.float32(0)
+ compute_2[cse_var_2 + 75] = T.float32(0)
+ compute_2[cse_var_2 + 76] = T.float32(0)
+ compute_2[cse_var_2 + 77] = T.float32(0)
+ compute_2[cse_var_2 + 78] = T.float32(0)
+ compute_2[cse_var_2 + 79] = T.float32(0)
+ compute_2[cse_var_2 + 96] = T.float32(0)
+ compute_2[cse_var_2 + 97] = T.float32(0)
+ compute_2[cse_var_2 + 98] = T.float32(0)
+ compute_2[cse_var_2 + 99] = T.float32(0)
+ compute_2[cse_var_2 + 100] = T.float32(0)
+ compute_2[cse_var_2 + 101] = T.float32(0)
+ compute_2[cse_var_2 + 102] = T.float32(0)
+ compute_2[cse_var_2 + 103] = T.float32(0)
+ compute_2[cse_var_2 + 104] = T.float32(0)
+ compute_2[cse_var_2 + 105] = T.float32(0)
+ compute_2[cse_var_2 + 106] = T.float32(0)
+ compute_2[cse_var_2 + 107] = T.float32(0)
+ compute_2[cse_var_2 + 108] = T.float32(0)
+ compute_2[cse_var_2 + 109] = T.float32(0)
+ compute_2[cse_var_2 + 110] = T.float32(0)
+ compute_2[cse_var_2 + 111] = T.float32(0)
+ compute_2[cse_var_2 + 128] = T.float32(0)
+ compute_2[cse_var_2 + 129] = T.float32(0)
+ compute_2[cse_var_2 + 130] = T.float32(0)
+ compute_2[cse_var_2 + 131] = T.float32(0)
+ compute_2[cse_var_2 + 132] = T.float32(0)
+ compute_2[cse_var_2 + 133] = T.float32(0)
+ compute_2[cse_var_2 + 134] = T.float32(0)
+ compute_2[cse_var_2 + 135] = T.float32(0)
+ compute_2[cse_var_2 + 136] = T.float32(0)
+ compute_2[cse_var_2 + 137] = T.float32(0)
+ compute_2[cse_var_2 + 138] = T.float32(0)
+ compute_2[cse_var_2 + 139] = T.float32(0)
+ compute_2[cse_var_2 + 140] = T.float32(0)
+ compute_2[cse_var_2 + 141] = T.float32(0)
+ compute_2[cse_var_2 + 142] = T.float32(0)
+ compute_2[cse_var_2 + 143] = T.float32(0)
+ compute_2[cse_var_2 + 160] = T.float32(0)
+ compute_2[cse_var_2 + 161] = T.float32(0)
+ compute_2[cse_var_2 + 162] = T.float32(0)
+ compute_2[cse_var_2 + 163] = T.float32(0)
+ compute_2[cse_var_2 + 164] = T.float32(0)
+ compute_2[cse_var_2 + 165] = T.float32(0)
+ compute_2[cse_var_2 + 166] = T.float32(0)
+ compute_2[cse_var_2 + 167] = T.float32(0)
+ compute_2[cse_var_2 + 168] = T.float32(0)
+ compute_2[cse_var_2 + 169] = T.float32(0)
+ compute_2[cse_var_2 + 170] = T.float32(0)
+ compute_2[cse_var_2 + 171] = T.float32(0)
+ compute_2[cse_var_2 + 172] = T.float32(0)
+ compute_2[cse_var_2 + 173] = T.float32(0)
+ compute_2[cse_var_2 + 174] = T.float32(0)
+ compute_2[cse_var_2 + 175] = T.float32(0)
+ compute_2[cse_var_2 + 192] = T.float32(0)
+ compute_2[cse_var_2 + 193] = T.float32(0)
+ compute_2[cse_var_2 + 194] = T.float32(0)
+ compute_2[cse_var_2 + 195] = T.float32(0)
+ compute_2[cse_var_2 + 196] = T.float32(0)
+ compute_2[cse_var_2 + 197] = T.float32(0)
+ compute_2[cse_var_2 + 198] = T.float32(0)
+ compute_2[cse_var_2 + 199] = T.float32(0)
+ compute_2[cse_var_2 + 200] = T.float32(0)
+ compute_2[cse_var_2 + 201] = T.float32(0)
+ compute_2[cse_var_2 + 202] = T.float32(0)
+ compute_2[cse_var_2 + 203] = T.float32(0)
+ compute_2[cse_var_2 + 204] = T.float32(0)
+ compute_2[cse_var_2 + 205] = T.float32(0)
+ compute_2[cse_var_2 + 206] = T.float32(0)
+ compute_2[cse_var_2 + 207] = T.float32(0)
+ compute_2[cse_var_2 + 224] = T.float32(0)
+ compute_2[cse_var_2 + 225] = T.float32(0)
+ compute_2[cse_var_2 + 226] = T.float32(0)
+ compute_2[cse_var_2 + 227] = T.float32(0)
+ compute_2[cse_var_2 + 228] = T.float32(0)
+ compute_2[cse_var_2 + 229] = T.float32(0)
+ compute_2[cse_var_2 + 230] = T.float32(0)
+ compute_2[cse_var_2 + 231] = T.float32(0)
+ compute_2[cse_var_2 + 232] = T.float32(0)
+ compute_2[cse_var_2 + 233] = T.float32(0)
+ compute_2[cse_var_2 + 234] = T.float32(0)
+ compute_2[cse_var_2 + 235] = T.float32(0)
+ compute_2[cse_var_2 + 236] = T.float32(0)
+ compute_2[cse_var_2 + 237] = T.float32(0)
+ compute_2[cse_var_2 + 238] = T.float32(0)
+ compute_2[cse_var_2 + 239] = T.float32(0)
+ for elem_idx in range(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_3: T.int32 = i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner
- cse_var_2: T.int32 = i_inner * 32 + nb_j_inner * 16 + j
+ cse_var_131: T.int32 = elem_idx * 16
+ cse_var_130: T.int32 = cse_var_2 + 99
+ cse_var_129: T.int32 = cse_var_2 + 98
+ cse_var_128: T.int32 = cse_var_2 + 97
+ cse_var_127: T.int32 = cse_var_2 + 96
+ cse_var_126: T.int32 = cse_var_2 + 9
+ cse_var_125: T.int32 = cse_var_2 + 8
+ cse_var_124: T.int32 = cse_var_2 + 79
+ cse_var_123: T.int32 = cse_var_2 + 78
+ cse_var_122: T.int32 = cse_var_2 + 77
+ cse_var_121: T.int32 = cse_var_2 + 76
+ cse_var_120: T.int32 = cse_var_2 + 75
+ cse_var_119: T.int32 = cse_var_2 + 74
+ cse_var_118: T.int32 = cse_var_2 + 73
+ cse_var_117: T.int32 = cse_var_2 + 72
+ cse_var_116: T.int32 = cse_var_2 + 71
+ cse_var_115: T.int32 = cse_var_2 + 70
+ cse_var_114: T.int32 = cse_var_2 + 7
+ cse_var_113: T.int32 = cse_var_2 + 69
+ cse_var_112: T.int32 = cse_var_2 + 68
+ cse_var_111: T.int32 = cse_var_2 + 67
+ cse_var_110: T.int32 = cse_var_2 + 66
+ cse_var_109: T.int32 = cse_var_2 + 65
+ cse_var_108: T.int32 = cse_var_2 + 64
+ cse_var_107: T.int32 = cse_var_2 + 6
+ cse_var_106: T.int32 = cse_var_2 + 5
+ cse_var_105: T.int32 = cse_var_2 + 47
+ cse_var_104: T.int32 = cse_var_2 + 46
+ cse_var_103: T.int32 = cse_var_2 + 45
+ cse_var_102: T.int32 = cse_var_2 + 44
+ cse_var_101: T.int32 = cse_var_2 + 43
+ cse_var_100: T.int32 = cse_var_2 + 42
+ cse_var_99: T.int32 = cse_var_2 + 41
+ cse_var_98: T.int32 = cse_var_2 + 40
+ cse_var_97: T.int32 = cse_var_2 + 4
+ cse_var_96: T.int32 = cse_var_2 + 39
+ cse_var_95: T.int32 = cse_var_2 + 38
+ cse_var_94: T.int32 = cse_var_2 + 37
+ cse_var_93: T.int32 = cse_var_2 + 36
+ cse_var_92: T.int32 = cse_var_2 + 35
+ cse_var_91: T.int32 = cse_var_2 + 34
+ cse_var_90: T.int32 = cse_var_2 + 33
+ cse_var_89: T.int32 = cse_var_2 + 32
+ cse_var_88: T.int32 = cse_var_2 + 3
+ cse_var_87: T.int32 = cse_var_2 + 239
+ cse_var_86: T.int32 = cse_var_2 + 238
+ cse_var_85: T.int32 = cse_var_2 + 237
+ cse_var_84: T.int32 = cse_var_2 + 236
+ cse_var_83: T.int32 = cse_var_2 + 235
+ cse_var_82: T.int32 = cse_var_2 + 234
+ cse_var_81: T.int32 = cse_var_2 + 233
+ cse_var_80: T.int32 = cse_var_2 + 232
+ cse_var_79: T.int32 = cse_var_2 + 231
+ cse_var_78: T.int32 = cse_var_2 + 230
+ cse_var_77: T.int32 = cse_var_2 + 229
+ cse_var_76: T.int32 = cse_var_2 + 228
+ cse_var_75: T.int32 = cse_var_2 + 227
+ cse_var_74: T.int32 = cse_var_2 + 226
+ cse_var_73: T.int32 = cse_var_2 + 225
+ cse_var_72: T.int32 = cse_var_2 + 224
+ cse_var_71: T.int32 = cse_var_2 + 207
+ cse_var_70: T.int32 = cse_var_2 + 206
+ cse_var_69: T.int32 = cse_var_2 + 205
+ cse_var_68: T.int32 = cse_var_2 + 204
+ cse_var_67: T.int32 = cse_var_2 + 203
+ cse_var_66: T.int32 = cse_var_2 + 202
+ cse_var_65: T.int32 = cse_var_2 + 201
+ cse_var_64: T.int32 = cse_var_2 + 200
+ cse_var_63: T.int32 = cse_var_2 + 2
+ cse_var_62: T.int32 = cse_var_2 + 199
+ cse_var_61: T.int32 = cse_var_2 + 198
+ cse_var_60: T.int32 = cse_var_2 + 197
+ cse_var_59: T.int32 = cse_var_2 + 196
+ cse_var_58: T.int32 = cse_var_2 + 195
+ cse_var_57: T.int32 = cse_var_2 + 194
+ cse_var_56: T.int32 = cse_var_2 + 193
+ cse_var_55: T.int32 = cse_var_2 + 192
+ cse_var_54: T.int32 = cse_var_2 + 175
+ cse_var_53: T.int32 = cse_var_2 + 174
+ cse_var_52: T.int32 = cse_var_2 + 173
+ cse_var_51: T.int32 = cse_var_2 + 172
+ cse_var_50: T.int32 = cse_var_2 + 171
+ cse_var_49: T.int32 = cse_var_2 + 170
+ cse_var_48: T.int32 = cse_var_2 + 169
+ cse_var_47: T.int32 = cse_var_2 + 168
+ cse_var_46: T.int32 = cse_var_2 + 167
+ cse_var_45: T.int32 = cse_var_2 + 166
+ cse_var_44: T.int32 = cse_var_2 + 165
+ cse_var_43: T.int32 = cse_var_2 + 164
+ cse_var_42: T.int32 = cse_var_2 + 163
+ cse_var_41: T.int32 = cse_var_2 + 162
+ cse_var_40: T.int32 = cse_var_2 + 161
+ cse_var_39: T.int32 = cse_var_2 + 160
+ cse_var_38: T.int32 = cse_var_2 + 15
+ cse_var_37: T.int32 = cse_var_2 + 143
+ cse_var_36: T.int32 = cse_var_2 + 142
+ cse_var_35: T.int32 = cse_var_2 + 141
+ cse_var_34: T.int32 = cse_var_2 + 140
+ cse_var_33: T.int32 = cse_var_2 + 14
+ cse_var_32: T.int32 = cse_var_2 + 139
+ cse_var_31: T.int32 = cse_var_2 + 138
+ cse_var_30: T.int32 = cse_var_2 + 137
+ cse_var_29: T.int32 = cse_var_2 + 136
+ cse_var_28: T.int32 = cse_var_2 + 135
+ cse_var_27: T.int32 = cse_var_2 + 134
+ cse_var_26: T.int32 = cse_var_2 + 133
+ cse_var_25: T.int32 = cse_var_2 + 132
+ cse_var_24: T.int32 = cse_var_2 + 131
+ cse_var_23: T.int32 = cse_var_2 + 130
+ cse_var_22: T.int32 = cse_var_2 + 13
+ cse_var_21: T.int32 = cse_var_2 + 129
+ cse_var_20: T.int32 = cse_var_2 + 128
+ cse_var_19: T.int32 = cse_var_2 + 12
+ cse_var_18: T.int32 = cse_var_2 + 111
+ cse_var_17: T.int32 = cse_var_2 + 110
+ cse_var_16: T.int32 = cse_var_2 + 11
+ cse_var_15: T.int32 = cse_var_2 + 109
+ cse_var_14: T.int32 = cse_var_2 + 108
+ cse_var_13: T.int32 = cse_var_2 + 107
+ cse_var_12: T.int32 = cse_var_2 + 106
+ cse_var_11: T.int32 = cse_var_2 + 105
+ cse_var_10: T.int32 = cse_var_2 + 104
+ cse_var_9: T.int32 = cse_var_2 + 103
+ cse_var_8: T.int32 = cse_var_2 + 102
+ cse_var_7: T.int32 = cse_var_2 + 101
+ cse_var_6: T.int32 = cse_var_2 + 100
+ cse_var_5: T.int32 = cse_var_2 + 10
+ cse_var_4: T.int32 = cse_var_2 + 1
+ cse_var_3: T.int32 = i0_outer_i1_outer_fused // 16 * 4096 + i_outer_inner * 2048
placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
placeholder_7 = T.Buffer((32768,), data=placeholder.data)
placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
- compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 16 * 8192 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- for i0_inner, i1_inner in T.grid(32, 32):
- cse_var_4: T.int32 = i0_outer_i1_outer_fused // 16 * 16384 + i0_inner * 512 + i0_outer_i1_outer_fused % 16 * 32 + i1_inner
+ compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_63] = compute_2[cse_var_63] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_88] = compute_2[cse_var_88] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_97] = compute_2[cse_var_97] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_106] = compute_2[cse_var_106] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_107] = compute_2[cse_var_107] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_114] = compute_2[cse_var_114] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_125] = compute_2[cse_var_125] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_126] = compute_2[cse_var_126] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_19] = compute_2[cse_var_19] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_22] = compute_2[cse_var_22] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_33] = compute_2[cse_var_33] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_38] = compute_2[cse_var_38] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_89] = compute_2[cse_var_89] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_90] = compute_2[cse_var_90] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_91] = compute_2[cse_var_91] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_92] = compute_2[cse_var_92] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_93] = compute_2[cse_var_93] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_94] = compute_2[cse_var_94] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_95] = compute_2[cse_var_95] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_96] = compute_2[cse_var_96] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_98] = compute_2[cse_var_98] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_99] = compute_2[cse_var_99] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_100] = compute_2[cse_var_100] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_101] = compute_2[cse_var_101] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_102] = compute_2[cse_var_102] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_103] = compute_2[cse_var_103] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_104] = compute_2[cse_var_104] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_105] = compute_2[cse_var_105] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_108] = compute_2[cse_var_108] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_109] = compute_2[cse_var_109] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_110] = compute_2[cse_var_110] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_111] = compute_2[cse_var_111] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_112] = compute_2[cse_var_112] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_113] = compute_2[cse_var_113] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_115] = compute_2[cse_var_115] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_116] = compute_2[cse_var_116] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_117] = compute_2[cse_var_117] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_118] = compute_2[cse_var_118] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_119] = compute_2[cse_var_119] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_120] = compute_2[cse_var_120] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_121] = compute_2[cse_var_121] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_122] = compute_2[cse_var_122] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_123] = compute_2[cse_var_123] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_124] = compute_2[cse_var_124] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_127] = compute_2[cse_var_127] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_128] = compute_2[cse_var_128] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_129] = compute_2[cse_var_129] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_130] = compute_2[cse_var_130] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_18] = compute_2[cse_var_18] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_20] = compute_2[cse_var_20] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_21] = compute_2[cse_var_21] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_23] = compute_2[cse_var_23] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_24] = compute_2[cse_var_24] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_25] = compute_2[cse_var_25] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_26] = compute_2[cse_var_26] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_27] = compute_2[cse_var_27] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_28] = compute_2[cse_var_28] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_29] = compute_2[cse_var_29] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_30] = compute_2[cse_var_30] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_31] = compute_2[cse_var_31] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_32] = compute_2[cse_var_32] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_34] = compute_2[cse_var_34] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_35] = compute_2[cse_var_35] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_36] = compute_2[cse_var_36] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_37] = compute_2[cse_var_37] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_39] = compute_2[cse_var_39] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_40] = compute_2[cse_var_40] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_41] = compute_2[cse_var_41] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_42] = compute_2[cse_var_42] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_43] = compute_2[cse_var_43] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_44] = compute_2[cse_var_44] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_45] = compute_2[cse_var_45] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_46] = compute_2[cse_var_46] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_47] = compute_2[cse_var_47] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_48] = compute_2[cse_var_48] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_49] = compute_2[cse_var_49] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_50] = compute_2[cse_var_50] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_51] = compute_2[cse_var_51] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_52] = compute_2[cse_var_52] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_53] = compute_2[cse_var_53] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_54] = compute_2[cse_var_54] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_55] = compute_2[cse_var_55] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_56] = compute_2[cse_var_56] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_57] = compute_2[cse_var_57] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_58] = compute_2[cse_var_58] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_59] = compute_2[cse_var_59] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_60] = compute_2[cse_var_60] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_61] = compute_2[cse_var_61] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_62] = compute_2[cse_var_62] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_64] = compute_2[cse_var_64] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_65] = compute_2[cse_var_65] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_66] = compute_2[cse_var_66] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_67] = compute_2[cse_var_67] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_68] = compute_2[cse_var_68] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_69] = compute_2[cse_var_69] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_70] = compute_2[cse_var_70] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_71] = compute_2[cse_var_71] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_72] = compute_2[cse_var_72] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_73] = compute_2[cse_var_73] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_74] = compute_2[cse_var_74] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_75] = compute_2[cse_var_75] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_76] = compute_2[cse_var_76] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_77] = compute_2[cse_var_77] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_78] = compute_2[cse_var_78] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_79] = compute_2[cse_var_79] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_80] = compute_2[cse_var_80] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_81] = compute_2[cse_var_81] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_82] = compute_2[cse_var_82] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_83] = compute_2[cse_var_83] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_84] = compute_2[cse_var_84] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_85] = compute_2[cse_var_85] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_86] = compute_2[cse_var_86] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_87] = compute_2[cse_var_87] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ for i0_inner, i1_inner in T.grid(16, 32):
+ cse_var_132: T.int32 = i0_outer_i1_outer_fused // 16 * 8192 + i0_inner * 512 + i0_outer_i1_outer_fused % 16 * 32 + i1_inner
compute_3 = T.Buffer((65536,), data=compute.data)
placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
- compute_3[cse_var_4] = T.max(compute_2[i0_inner * 32 + i1_inner] + placeholder_5[cse_var_4], T.float32(0))
+ compute_3[cse_var_132] = T.max(compute_2[i0_inner * 32 + i1_inner] + placeholder_5[cse_var_132], T.float32(0))
@@ -458,7 +839,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 1.656 ms
+ Execution time of this operator: 2.751 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 48412967bf..8c2817b51e 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
@@ -5,16 +5,16 @@
Computation times
=================
-**00:47.212** total execution time for **how_to_tune_with_autotvm** files:
+**00:42.856** 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``) | 00:47.168 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:42.818 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.027 | 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_cuda.py` (``tune_relay_cuda.py``) | 00:00.006 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``) | 00:00.006 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``) | 00:00.005 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``) | 00:00.006 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``) | 00:00.005 | 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 866907ea29..68d687cba2 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
@@ -268,7 +268,8 @@ for this template
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ No: 1 GFLOPS: 338.11/338.11 result: MeasureResult(costs=(0.0006847002883435583,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0730724334716797, timestamp=1683281099.2884815) [('tile_f', [-1, 1, 32, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5490850
+ No: 2 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -390,8 +391,26 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5171664
- No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 32, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3149212
+ No: 3 GFLOPS: 0.00/338.11 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
+ return self.__get_result()
+ File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
+ raise self._exception
+ File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
+ result = self.fn(*self.args, **self.kwargs)
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 432, in <lambda>
+ worker = lambda *args: self._worker_run(*args)
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 401, in _worker_run
+ return proc.recv()
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 309, in recv
+ raise TimeoutError()
+ TimeoutError
+
+ [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3690178
+ No: 4 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -513,8 +532,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8050664
- No: 3 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3076143
+ No: 5 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -636,8 +655,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 64, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2898546
- No: 4 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1105001
+ No: 6 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -759,10 +778,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2015182
- No: 5 GFLOPS: 40.55/40.55 result: MeasureResult(costs=(0.005709265590909091,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.552736520767212, timestamp=1683267721.9495256) [('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5906714
- No: 6 GFLOPS: 40.48/40.55 result: MeasureResult(costs=(0.005719022111111111,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.378983497619629, timestamp=1683267722.848646) [('tile_f', [-1, 32, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4656104
- No: 7 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,614915
+ No: 7 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -884,8 +901,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5021600
- No: 8 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3932741
+ No: 8 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1007,9 +1024,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 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,10194376
- No: 9 GFLOPS: 7.35/40.55 result: MeasureResult(costs=(0.031504911249999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.7945895195007324, timestamp=1683267726.8854792) [('tile_f', [-1, 4, 2, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9293632
- No: 10 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 128, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,280991
+ No: 9 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1131,8 +1147,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 256, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8270339
- No: 11 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 128]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,307110
+ No: 10 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1254,8 +1270,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9261023
- No: 12 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 512, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1889589
+ No: 11 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1377,8 +1393,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5938384
- No: 13 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 64, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6259485
+ No: 12 GFLOPS: 40.04/338.11 result: MeasureResult(costs=(0.005781580037037037,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.5704667568206787, timestamp=1683281109.8871222) [('tile_f', [-1, 8, 16, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8784509
+ No: 13 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1500,8 +1517,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4508921
- No: 14 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4510715
+ No: 14 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1623,26 +1640,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6801698
- No: 15 GFLOPS: 0.00/40.55 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
- return self.__get_result()
- File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
- raise self._exception
- File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
- result = self.fn(*self.args, **self.kwargs)
- File "/workspace/python/tvm/contrib/popen_pool.py", line 432, in <lambda>
- worker = lambda *args: self._worker_run(*args)
- File "/workspace/python/tvm/contrib/popen_pool.py", line 401, in _worker_run
- return proc.recv()
- File "/workspace/python/tvm/contrib/popen_pool.py", line 309, in recv
- raise TimeoutError()
- TimeoutError
-
- [('tile_f', [-1, 1, 1, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10315496
- No: 16 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 16, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1251448
+ No: 15 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1764,8 +1763,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 256, 1, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,867963
- No: 17 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,697246
+ No: 16 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1887,8 +1886,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3906436
- No: 18 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2271861
+ No: 17 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2010,8 +2009,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3201082
- No: 19 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 64, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1408708
+ No: 18 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2133,8 +2132,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 32, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3960483
- No: 20 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7541027
+ No: 19 GFLOPS: 19.58/338.11 result: MeasureResult(costs=(0.011821615,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.6270549297332764, timestamp=1683281112.9016883) [('tile_f', [-1, 8, 1, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3257323
+ No: 20 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2256,7 +2256,7 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2128011
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4826081
@@ -2311,9 +2311,9 @@ and measure running time.
Finish loading 20 records
Best config:
- [('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5906714
+ [('tile_f', [-1, 1, 32, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5490850
Finish loading 20 records
- Time cost of this operator: 0.006087
+ Time cost of this operator: 0.001033
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 6cfaf5b8b9..4633785546 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
@@ -360,10 +360,10 @@ 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 318.8 98.738 (1, 2, 10, 10, 3) 2 1 [318.8]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.108 0.963 (1, 6, 10, 10) 1 1 [3.108]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.965 0.299 (1, 1, 10, 10, 3) 1 1 [0.965]
- Total_time - 322.873 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 316.4 98.752 (1, 2, 10, 10, 3) 2 1 [316.4]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.045 0.95 (1, 6, 10, 10) 1 1 [3.045]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.953 0.298 (1, 1, 10, 10, 3) 1 1 [0.953]
+ Total_time - 320.398 - - - - -
@@ -428,10 +428,10 @@ 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 103.4 97.521 (1, 6, 10, 10, 1) 2 1 [103.4]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.773 1.672 (1, 6, 10, 10) 1 1 [1.773]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.855 0.807 (1, 3, 10, 10, 1) 1 1 [0.855]
- Total_time - 106.029 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 103.0 97.484 (1, 6, 10, 10, 1) 2 1 [103.0]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.781 1.686 (1, 6, 10, 10) 1 1 [1.781]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.878 0.831 (1, 3, 10, 10, 1) 1 1 [0.878]
+ Total_time - 105.659 - - - - -
@@ -439,7 +439,7 @@ Timing the tuned program
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 35.988 seconds)
+ **Total running time of the script:** ( 1 minutes 29.610 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_autotune.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
index 546a957dee..76112ab08e 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
@@ -118,7 +118,7 @@ download a cat image and preprocess it to use as the model input.
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/ao/quantization/utils.py:281: UserWarning: must run observer before calling calculate_qparams. Returning default values.
"must run observer before calling calculate_qparams. " +
Downloading: "https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
-
0%| | 0.00/3.42M [00:00<?, ?B/s]
100%|##########| 3.42M/3.42M [00:00<00:00, 81.5MB/s]
+
0%| | 0.00/3.42M [00:00<?, ?B/s]
100%|##########| 3.42M/3.42M [00:00<00:00, 39.7MB/s]
/workspace/python/tvm/relay/frontend/pytorch_utils.py:47: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
return LooseVersion(torch_ver) > ver
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
@@ -324,7 +324,7 @@ Look up prediction top 1 index in 1000 class synset.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 26.593 seconds)
+ **Total running time of the script:** ( 1 minutes 23.351 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_pytorch.py:
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 51a04256b2..f70d112c98 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
@@ -217,7 +217,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
.. code-block:: none
- '/tmp/tmpt3mzqdty/images/random'
+ '/tmp/tmp9yxg_u8h/images/random'
@@ -308,7 +308,7 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. image-sg:: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
- :alt: [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]
+ :alt: [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0]
:srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
:class: sphx-glr-single-img
@@ -317,8 +317,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmpt3mzqdty/images/target contains 8144 images
- /tmp/tmpt3mzqdty/images/random contains 5000 images
+ /tmp/tmp9yxg_u8h/images/target contains 8144 images
+ /tmp/tmp9yxg_u8h/images/random contains 5000 images
@@ -493,13 +493,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 43s - loss: 0.2601 - accuracy: 0.9158 - val_loss: 0.1228 - val_accuracy: 0.9562 - 43s/epoch - 132ms/step
+ 328/328 - 41s - loss: 0.2093 - accuracy: 0.9268 - val_loss: 0.1127 - val_accuracy: 0.9569 - 41s/epoch - 125ms/step
Epoch 2/3
- 328/328 - 37s - loss: 0.0967 - accuracy: 0.9618 - val_loss: 0.1026 - val_accuracy: 0.9649 - 37s/epoch - 111ms/step
+ 328/328 - 35s - loss: 0.1037 - accuracy: 0.9624 - val_loss: 0.0929 - val_accuracy: 0.9653 - 35s/epoch - 108ms/step
Epoch 3/3
- 328/328 - 36s - loss: 0.0631 - accuracy: 0.9779 - val_loss: 0.1065 - val_accuracy: 0.9611 - 36s/epoch - 111ms/step
+ 328/328 - 35s - loss: 0.0718 - accuracy: 0.9727 - val_loss: 0.0874 - val_accuracy: 0.9694 - 35s/epoch - 107ms/step
- <keras.callbacks.History object at 0x7fb2053a1fd0>
+ <keras.callbacks.History object at 0x7fa9533a0b50>
@@ -860,7 +860,7 @@ 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:** ( 4 minutes 51.119 seconds)
+ **Total running time of the script:** ( 4 minutes 37.968 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 6e0b1b08d7..a4e2e6b121 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
@@ -5,20 +5,20 @@
Computation times
=================
-**08:22.699** total execution time for **how_to_work_with_microtvm** files:
+**07:58.155** total execution time for **how_to_work_with_microtvm** files:
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:51.119 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:37.968 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 01:35.988 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 01:29.610 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``) | 01:26.593 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``) | 01:23.351 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:11.533 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:10.862 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:09.159 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:08.646 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:08.307 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:07.718 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``) | 00:00.000 | 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 f43a76d46b..6a536a6063 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
=================
-**00:42.141** total execution time for **how_to_work_with_relay** files:
+**00:38.817** 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``) | 00:36.987 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:34.016 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:03.312 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:03.093 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.834 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.702 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``) | 00:00.008 | 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 f884ab807e..d81c370a87 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
@@ -278,7 +278,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
.. code-block:: none
- <function my_cuda_math_rule at 0x7fae526ed950>
+ <function my_cuda_math_rule at 0x7fa59e48a950>
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 6717c964e3..34ae8ebc92 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
=================
-**00:09.573** total execution time for **how_to_work_with_schedules** files:
+**00:07.020** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:06.494 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:04.149 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.375 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.268 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.703 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.664 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.672 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.648 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.153 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.137 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.073 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.065 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.067 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.058 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.036 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.031 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
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 354911cdaf..9e250e1ae2 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
=================
-**00:36.052** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:33.242** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:36.044 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:33.235 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``) | 00:00.008 | 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 68e9c37668..dab5b1bf5b 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -293,7 +293,7 @@ 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 37.67s!
+ resnet18_v1 inference graph built in 35.90s!
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 30ad96c05f..f729d9ca5c 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -337,7 +337,7 @@ The compilation steps are:
/workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 26.70s!
+ yolov3-tiny inference graph built in 24.39s!
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 6fea3b46bd..372fd71812 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
=================
-**01:49.228** total execution time for **topic_vta_tutorials_frontend** files:
+**01:44.850** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:54.857 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:52.980 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:54.372 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:51.870 | 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 c423f225a9..e4b79ec02a 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
=================
-**00:03.358** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.320** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.823 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.781 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.535 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.539 | 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 67217af58f..f00ac232a0 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
=================
-**00:00.910** total execution time for **topic_vta_tutorials** files:
+**00:00.917** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.470 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.472 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.441 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.445 | 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 66470b2042..aca24a21bc 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -318,7 +318,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 95.179 ms
+ Execution time of this operator: 94.060 ms
@@ -434,7 +434,7 @@ operations.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 39.679 seconds)
+ **Total running time of the script:** ( 1 minutes 34.545 seconds)
.. _sphx_glr_download_tutorial_auto_scheduler_matmul_x86.py:
diff --git a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
index a60860c2df..289e6b83fb 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -454,16 +454,16 @@ reduce variance, we take 5 measurements and average them.
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 3.38/3.38 result: MeasureResult(costs=(0.079426669,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5549054145812988, timestamp=1683265863.1309867) [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
- No: 2 GFLOPS: 1.32/3.38 result: MeasureResult(costs=(0.2026048132,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.494455575942993, timestamp=1683265868.1066635) [('tile_y', [-1, 2]), ('tile_x', [-1, 1])],None,1
- No: 3 GFLOPS: 10.50/10.50 result: MeasureResult(costs=(0.025575716600000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6876952648162842, timestamp=1683265868.7841988) [('tile_y', [-1, 128]), ('tile_x', [-1, 32])],None,57
- No: 4 GFLOPS: 11.31/11.31 result: MeasureResult(costs=(0.0237292448,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6627187728881836, timestamp=1683265870.8939292) [('tile_y', [-1, 8]), ('tile_x', [-1, 512])],None,93
- No: 5 GFLOPS: 3.28/11.31 result: MeasureResult(costs=(0.0818135528,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5976903438568115, timestamp=1683265872.6470945) [('tile_y', [-1, 64]), ('tile_x', [-1, 8])],None,36
- No: 6 GFLOPS: 9.76/11.31 result: MeasureResult(costs=(0.0274978006,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7093219757080078, timestamp=1683265874.8304014) [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
- No: 7 GFLOPS: 12.30/12.30 result: MeasureResult(costs=(0.0218288464,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6348245143890381, timestamp=1683265875.455425) [('tile_y', [-1, 32]), ('tile_x', [-1, 128])],None,75
- No: 8 GFLOPS: 10.96/12.30 result: MeasureResult(costs=(0.0244883356,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7110598087310791, timestamp=1683265876.122472) [('tile_y', [-1, 16]), ('tile_x', [-1, 256])],None,84
- No: 9 GFLOPS: 4.06/12.30 result: MeasureResult(costs=(0.0660457318,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3017196655273438, timestamp=1683265877.5584345) [('tile_y', [-1, 4]), ('tile_x', [-1, 16])],None,42
- No: 10 GFLOPS: 12.95/12.95 result: MeasureResult(costs=(0.0207274876,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5882182121276855, timestamp=1683265878.158813) [('tile_y', [-1, 128]), ('tile_x', [-1, 64])],None,67
+ No: 1 GFLOPS: 11.52/11.52 result: MeasureResult(costs=(0.0233052484,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6609020233154297, timestamp=1683279342.7045352) [('tile_y', [-1, 8]), ('tile_x', [-1, 256])],None,83
+ No: 2 GFLOPS: 11.52/11.52 result: MeasureResult(costs=(0.0232966164,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.635930061340332, timestamp=1683279343.344799) [('tile_y', [-1, 64]), ('tile_x', [-1, 512])],None,96
+ No: 3 GFLOPS: 1.80/11.52 result: MeasureResult(costs=(0.1490323126,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.625932216644287, timestamp=1683279347.2974453) [('tile_y', [-1, 1]), ('tile_x', [-1, 4])],None,20
+ No: 4 GFLOPS: 1.55/11.52 result: MeasureResult(costs=(0.1727030786,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.0179309844970703, timestamp=1683279351.6501443) [('tile_y', [-1, 16]), ('tile_x', [-1, 1])],None,4
+ No: 5 GFLOPS: 10.35/11.52 result: MeasureResult(costs=(0.0259372634,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7117331027984619, timestamp=1683279352.4766116) [('tile_y', [-1, 4]), ('tile_x', [-1, 128])],None,72
+ No: 6 GFLOPS: 2.68/11.52 result: MeasureResult(costs=(0.10028091940000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8588132858276367, timestamp=1683279354.3398387) [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
+ No: 7 GFLOPS: 0.90/11.52 result: MeasureResult(costs=(0.299469356,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.050477027893066, timestamp=1683279360.7322223) [('tile_y', [-1, 64]), ('tile_x', [-1, 2])],None,16
+ No: 8 GFLOPS: 13.05/13.05 result: MeasureResult(costs=(0.0205636638,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5920419692993164, timestamp=1683279361.323545) [('tile_y', [-1, 256]), ('tile_x', [-1, 64])],None,68
+ No: 9 GFLOPS: 4.12/13.05 result: MeasureResult(costs=(0.0651398906,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.277179479598999, timestamp=1683279362.7167172) [('tile_y', [-1, 4]), ('tile_x', [-1, 16])],None,42
+ No: 10 GFLOPS: 11.48/13.05 result: MeasureResult(costs=(0.023376216999999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6081831455230713, timestamp=1683279363.3579586) [('tile_y', [-1, 4]), ('tile_x', [-1, 256])],None,82
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index ff70dc26c7..9edec2616a 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -311,7 +311,7 @@ standard deviation.
.. code-block:: none
- {'mean': 528.2606210899996, 'median': 527.0514347999949, 'std': 4.135171668509399}
+ {'mean': 516.5330643200014, 'median': 516.7005607999727, 'std': 2.309861836840107}
@@ -582,31 +582,30 @@ the tuning data to.
.. code-block:: none
-
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 10.96/ 21.64 GFLOPS | Progress: (4/20) | 13.60 s
[Task 1/25] Current/Best: 6.81/ 21.64 GFLOPS | Progress: (8/20) | 21.77 s
[Task 1/25] Current/Best: 11.21/ 21.64 GFLOPS | Progress: (12/20) | 24.66 s
[Task 1/25] Current/Best: 17.24/ 23.40 GFLOPS | Progress: (16/20) | 26.72 s
[Task 1/25] Current/Best: 6.86/ 23.40 GFLOPS | Progress: (20/20) | 29.17 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.75/ 14.60 GFLOPS | Progress: (4/20) | 5.66 s
[Task 2/25] Current/Best: 3.70/ 19.69 GFLOPS | Progress: (8/20) | 7.35 s
[Task 2/25] Current/Best: 3.74/ 19.69 GFLOPS | Progress: (12/20) | 9.09 s
[Task 2/25] Current/Best: 5.70/ 19.69 GFLOPS | Progress: (16/20) | 10.61 s
[Task 2/25] Current/Best: 9.45/ 19.69 GFLOPS | Progress: (20/20) | 12.53 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 18.00/ 21.83 GFLOPS | Progress: (4/20) | 6.13 s
[Task 3/25] Current/Best: 23.46/ 23.46 GFLOPS | Progress: (8/20) | 8.50 s
[Task 3/25] Current/Best: 17.33/ 23.46 GFLOPS | Progress: (12/20) | 11.33 s
[Task 3/25] Current/Best: 19.38/ 23.46 GFLOPS | Progress: (16/20) | 13.56 s
[Task 3/25] Current/Best: 6.82/ 23.46 GFLOPS | Progress: (20/20) | 16.41 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 13.44/ 17.85 GFLOPS | Progress: (4/20) | 5.66 s
[Task 4/25] Current/Best: 3.81/ 18.70 GFLOPS | Progress: (8/20) | 7.62 s
[Task 4/25] Current/Best: 11.04/ 19.97 GFLOPS | Progress: (12/20) | 9.94 s
[Task 4/25] Current/Best: 4.98/ 19.97 GFLOPS | Progress: (16/20) | 15.34 s
[Task 4/25] Current/Best: 5.65/ 19.97 GFLOPS | Progress: (20/20) | 18.98 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 15.22/ 17.53 GFLOPS | Progress: (4/20) | 5.04 s
[Task 5/25] Current/Best: 17.63/ 17.63 GFLOPS | Progress: (8/20) | 7.22 s
[Task 5/25] Current/Best: 9.37/ 17.63 GFLOPS | Progress: (12/20) | 9.51 s
[Task 5/25] Current/Best: 18.77/ 18.77 GFLOPS | Progress: (16/20) | 11.77 s
[Task 5/25] Current/Best: 7.98/ 18.77 GFLOPS | Progress: (20/20) | 14.31 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 11.99/ 14.18 GFLOPS | Progress: (4/20) | 6.26 s
[Task 6/25] Current/Best: 2.93/ 17.27 GFLOPS | Progress: (8/20) | 9.27 s
[Task 6/25] Current/Best: 4.64/ 17.27 GFLOPS | Progress: (12/20) | 12.57 s
[Task 6/25] Current/Best: 20.12/ 20.12 GFLOPS | Progress: (16/20) | 19.77 s
[Task 6/25] Current/Best: 9.26/ 20.12 GFLOPS | Progress: (20/20) | 22.93 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 5.71/ 18.99 GFLOPS | Progress: (4/20) | 6.16 s
[Task 7/25] Current/Best: 6.00/ 18.99 GFLOPS | Progress: (8/20) | 8.82 s
[Task 7/25] Current/Best: 12.41/ 18.99 GFLOPS | Progress: (12/20) | 12.33 s
[Task 7/25] Current/Best: 14.11/ 18.99 GFLOPS | Progress: (16/20) | 15.53 s
[Task 7/25] Current/Best: 7.55/ 18.99 GFLOPS | Progress: (20/20) | 18.10 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 12.80/ 13.95 GFLOPS | Progress: (4/20) | 6.33 s
[Task 8/25] Current/Best: 3.88/ 13.95 GFLOPS | Progress: (8/20) | 9.36 s
[Task 8/25] Current/Best: 10.92/ 15.13 GFLOPS | Progress: (12/20) | 13.10 s
[Task 8/25] Current/Best: 9.36/ 17.49 GFLOPS | Progress: (16/20) | 16.13 s
[Task 8/25] Current/Best: 11.77/ 19.42 GFLOPS | Progress: (20/20) | 22.29 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 6.92/ 18.98 GFLOPS | Progress: (4/20) | 5.41 s
[Task 9/25] Current/Best: 9.27/ 18.98 GFLOPS | Progress: (8/20) | 7.36 s
[Task 9/25] Current/Best: 11.22/ 18.98 GFLOPS | Progress: (12/20) | 10.21 s
[Task 9/25] Current/Best: 16.25/ 19.24 GFLOPS | Progress: (16/20) | 18.57 s
[Task 9/25] Current/Best: 11.33/ 19.24 GFLOPS | Progress: (20/20) | 24.44 s Done.
-
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 9.70/ 9.96 GFLOPS | Progress: (4/20) | 7.69 s
[Task 10/25] Current/Best: 13.47/ 15.91 GFLOPS | Progress: (8/20) | 9.60 s
[Task 10/25] Current/Best: 15.86/ 15.91 GFLOPS | Progress: (12/20) | 13.38 s
[Task 10/25] Current/Best: 5.91/ 22.26 GFLOPS | Progress: (16/20) | 15.05 s
[Task 10/25] Current/Best: 13.02/ 22.26 GFLOPS | Progress: (20/20) | 16.84 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 16.08/ 17.26 GFLOPS | Progress: (4/20) | 5.77 s
[Task 11/25] Current/Best: 10.16/ 18.29 GFLOPS | Progress: (8/20) | 8.97 s
[Task 11/25] Current/Best: 6.13/ 19.36 GFLOPS | Progress: (12/20) | 11.38 s
[Task 11/25] Current/Best: 10.77/ 19.36 GFLOPS | Progress: (16/20) | 13.95 s
[Task 11/25] Current/Best: 6.98/ 19.36 GFLOPS | Progress: (20/20) | 16.36 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 18.15/ 18.15 GFLOPS | Progress: (4/20) | 5.60 s
[Task 12/25] Current/Best: 9.54/ 18.15 GFLOPS | Progress: (8/20) | 8.20 s
[Task 12/25] Current/Best: 7.94/ 18.15 GFLOPS | Progress: (12/20) | 14.94 s
[Task 12/25] Current/Best: 10.62/ 18.15 GFLOPS | Progress: (16/20) | 17.80 s
[Task 12/25] Current/Best: 18.78/ 18.98 GFLOPS | Progress: (20/20) | 20.74 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 10.09/ 19.88 GFLOPS | Progress: (4/20) | 7.33 s
[Task 13/25] Current/Best: 18.21/ 19.88 GFLOPS | Progress: (8/20) | 9.42 s
[Task 13/25] Current/Best: 10.99/ 19.88 GFLOPS | Progress: (12/20) | 13.27 s
[Task 13/25] Current/Best: 11.96/ 19.88 GFLOPS | Progress: (16/20) | 17.41 s
[Task 13/25] Current/Best: 16.06/ 19.88 GFLOPS | Progress: (20/20) | 22.02 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 10.40/ 20.06 GFLOPS | Progress: (4/20) | 5.30 s
[Task 14/25] Current/Best: 11.95/ 20.06 GFLOPS | Progress: (8/20) | 7.97 s
[Task 14/25] Current/Best: 9.36/ 20.06 GFLOPS | Progress: (12/20) | 13.01 s
[Task 14/25] Current/Best: 13.73/ 20.06 GFLOPS | Progress: (16/20) | 16.10 s
[Task 14/25] Current/Best: 2.37/ 20.06 GFLOPS | Progress: (20/20) | 19.82 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 13.68/ 20.30 GFLOPS | Progress: (4/20) | 5.98 s
[Task 15/25] Current/Best: 6.79/ 20.30 GFLOPS | Progress: (8/20) | 8.83 s Done.
-
[Task 15/25] Current/Best: 10.76/ 20.30 GFLOPS | Progress: (12/20) | 14.33 s
[Task 15/25] Current/Best: 12.69/ 20.30 GFLOPS | Progress: (16/20) | 16.69 s
[Task 15/25] Current/Best: 12.59/ 20.30 GFLOPS | Progress: (20/20) | 19.94 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 15.38/ 18.07 GFLOPS | Progress: (4/20) | 5.07 s
[Task 16/25] Current/Best: 18.46/ 18.46 GFLOPS | Progress: (8/20) | 6.73 s
[Task 16/25] Current/Best: 10.57/ 18.84 GFLOPS | Progress: (12/20) | 8.97 s
[Task 16/25] Current/Best: 12.33/ 18.84 GFLOPS | Progress: (16/20) | 12.36 s
[Task 16/25] Current/Best: 6.49/ 19.45 GFLOPS | Progress: (20/20) | 14.09 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 12.17/ 21.49 GFLOPS | Progress: (4/20) | 5.93 s
[Task 17/25] Current/Best: 15.71/ 21.49 GFLOPS | Progress: (8/20) | 10.05 s
[Task 17/25] Current/Best: 20.60/ 21.49 GFLOPS | Progress: (12/20) | 13.01 s
[Task 17/25] Current/Best: 17.82/ 22.37 GFLOPS | Progress: (16/20) | 15.13 s
[Task 17/25] Current/Best: 11.79/ 23.06 GFLOPS | Progress: (20/20) | 18.70 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 9.25/ 19.00 GFLOPS | Progress: (4/20) | 5.81 s
[Task 18/25] Current/Best: 9.04/ 19.00 GFLOPS | Progress: (8/20) | 13.40 s
[Task 18/25] Current/Best: 18.31/ 19.00 GFLOPS | Progress: (12/20) | 16.01 s
[Task 18/25] Current/Best: 7.38/ 19.00 GFLOPS | Progress: (16/20) | 21.09 s
[Task 18/25] Current/Best: 10.02/ 19.00 GFLOPS | Progress: (20/20) | 25.21 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 16.90/ 16.90 GFLOPS | Progress: (4/20) | 6.08 s
[Task 19/25] Current/Best: 12.61/ 18.18 GFLOPS | Progress: (8/20) | 9.23 s
[Task 19/25] Current/Best: 1.55/ 18.18 GFLOPS | Progress: (12/20) | 14.39 s
[Task 19/25] Current/Best: 8.96/ 21.19 GFLOPS | Progress: (16/20) | 17.26 s
[Task 19/25] Current/Best: 8.47/ 21.19 GFLOPS | Progress: (20/20) | 20.77 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 18.15/ 18.15 GFLOPS | Progress: (4/20) | 4.45 s
[Task 20/25] Current/Best: 5.03/ 18.15 GFLOPS | Progress: (8/20) | 9.45 s
[Task 20/25] Current/Best: 16.97/ 18.15 GFLOPS | Progress: (12/20) | 11.99 s
[Task 20/25] Current/Best: 10.36/ 18.15 GFLOPS | Progress: (16/20) | 16.46 s
[Task 20/25] Current/Best: 2.07/ 18.15 GFLOPS | Progress: (20/20) | 20.67 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 14.93/ 14.93 GFLOPS | Progress: (4/20) | 6.19 s Done.
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 23.79/ 23.79 GFLOPS | Progress: (4/20) | 12.18 s
[Task 1/25] Current/Best: 10.92/ 23.79 GFLOPS | Progress: (8/20) | 16.10 s
[Task 1/25] Current/Best: 12.97/ 23.79 GFLOPS | Progress: (12/20) | 20.57 s
[Task 1/25] Current/Best: 23.65/ 23.79 GFLOPS | Progress: (16/20) | 23.78 s
[Task 1/25] Current/Best: 15.44/ 23.79 GFLOPS | Progress: (20/20) | 26.49 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 5.77/ 6.48 GFLOPS | Progress: (4/20) | 4.43 s
[Task 2/25] Current/Best: 6.58/ 14.66 GFLOPS | Progress: (8/20) | 5.91 s
[Task 2/25] Current/Best: 19.86/ 20.57 GFLOPS | Progress: (12/20) | 7.93 s
[Task 2/25] Current/Best: 16.18/ 20.57 GFLOPS | Progress: (16/20) | 9.43 s
[Task 2/25] Current/Best: 13.02/ 20.57 GFLOPS | Progress: (20/20) | 11.21 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 11.61/ 14.90 GFLOPS | Progress: (4/20) | 6.76 s
[Task 3/25] Current/Best: 14.93/ 14.93 GFLOPS | Progress: (8/20) | 9.24 s
[Task 3/25] Current/Best: 11.66/ 19.90 GFLOPS | Progress: (12/20) | 11.26 s
[Task 3/25] Current/Best: 12.39/ 19.90 GFLOPS | Progress: (16/20) | 13.43 s
[Task 3/25] Current/Best: 13.42/ 19.90 GFLOPS | Progress: (20/20) | 15.74 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 17.73/ 19.62 GFLOPS | Progress: (4/20) | 5.10 s
[Task 4/25] Current/Best: 15.49/ 19.62 GFLOPS | Progress: (8/20) | 11.35 s
[Task 4/25] Current/Best: 10.88/ 19.62 GFLOPS | Progress: (12/20) | 13.60 s
[Task 4/25] Current/Best: 22.72/ 22.72 GFLOPS | Progress: (16/20) | 15.94 s
[Task 4/25] Current/Best: 17.92/ 22.72 GFLOPS | Progress: (20/20) | 17.73 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 16.26/ 16.26 GFLOPS | Progress: (4/20) | 5.66 s
[Task 5/25] Current/Best: 17.77/ 23.68 GFLOPS | Progress: (8/20) | 7.89 s
[Task 5/25] Current/Best: 8.54/ 23.68 GFLOPS | Progress: (12/20) | 11.08 s
[Task 5/25] Current/Best: 15.45/ 23.68 GFLOPS | Progress: (16/20) | 12.97 s
[Task 5/25] Current/Best: 13.38/ 23.68 GFLOPS | Progress: (20/20) | 15.44 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 3.06/ 15.96 GFLOPS | Progress: (4/20) | 5.97 s
[Task 6/25] Current/Best: 17.53/ 22.12 GFLOPS | Progress: (8/20) | 7.90 s
[Task 6/25] Current/Best: 10.69/ 22.12 GFLOPS | Progress: (12/20) | 11.09 s
[Task 6/25] Current/Best: 12.54/ 22.12 GFLOPS | Progress: (16/20) | 13.85 s
[Task 6/25] Current/Best: 16.72/ 22.12 GFLOPS | Progress: (20/20) | 17.52 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 12.69/ 19.74 GFLOPS | Progress: (4/20) | 5.24 s
[Task 7/25] Current/Best: 14.00/ 19.74 GFLOPS | Progress: (8/20) | 7.66 s
[Task 7/25] Current/Best: 16.05/ 19.74 GFLOPS | Progress: (12/20) | 10.92 s
[Task 7/25] Current/Best: 21.52/ 22.23 GFLOPS | Progress: (16/20) | 12.91 s
[Task 7/25] Current/Best: 16.89/ 23.22 GFLOPS | Progress: (20/20) | 15.62 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 11.54/ 12.04 GFLOPS | Progress: (4/20) | 7.88 s
[Task 8/25] Current/Best: 10.51/ 12.66 GFLOPS | Progress: (8/20) | 10.60 s
[Task 8/25] Current/Best: 20.53/ 20.53 GFLOPS | Progress: (12/20) | 12.76 s
[Task 8/25] Current/Best: 11.25/ 20.53 GFLOPS | Progress: (16/20) | 21.46 s
[Task 8/25] Current/Best: 10.09/ 20.53 GFLOPS | Progress: (20/20) | 27.48 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 13.09/ 22.93 GFLOPS | Progress: (4/20) | 4.98 s
[Task 9/25] Current/Best: 4.54/ 22.93 GFLOPS | Progress: (8/20) | 12.76 s
[Task 9/25] Current/Best: 12.72/ 22.93 GFLOPS | Progress: (12/20) | 14.82 s
[Task 9/25] Current/Best: 19.20/ 22.93 GFLOPS | Progress: (16/20) | 21.86 s
[Task 9/25] Current/Best: 8.08/ 22.93 GFLOPS | Progress: (20/20) | 25.04 s Done.
+
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 18.10/ 18.10 GFLOPS | Progress: (4/20) | 4.68 s
[Task 10/25] Current/Best: 14.11/ 18.10 GFLOPS | Progress: (8/20) | 8.10 s
[Task 10/25] Current/Best: 10.50/ 18.10 GFLOPS | Progress: (12/20) | 10.84 s
[Task 10/25] Current/Best: 4.34/ 18.10 GFLOPS | Progress: (16/20) | 13.19 s
[Task 10/25] Current/Best: 8.95/ 18.10 GFLOPS | Progress: (20/20) | 17.40 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 17.25/ 23.52 GFLOPS | Progress: (4/20) | 5.47 s
[Task 11/25] Current/Best: 9.74/ 23.52 GFLOPS | Progress: (8/20) | 7.99 s
[Task 11/25] Current/Best: 9.27/ 23.52 GFLOPS | Progress: (12/20) | 12.19 s
[Task 11/25] Current/Best: 5.94/ 23.52 GFLOPS | Progress: (16/20) | 16.70 s
[Task 11/25] Current/Best: 17.17/ 23.52 GFLOPS | Progress: (20/20) | 19.16 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 3.72/ 18.92 GFLOPS | Progress: (4/20) | 5.38 s
[Task 12/25] Current/Best: 6.06/ 18.92 GFLOPS | Progress: (8/20) | 8.30 s
[Task 12/25] Current/Best: 6.62/ 19.73 GFLOPS | Progress: (12/20) | 11.16 s
[Task 12/25] Current/Best: 13.55/ 19.73 GFLOPS | Progress: (16/20) | 13.83 s
[Task 12/25] Current/Best: 3.75/ 19.73 GFLOPS | Progress: (20/20) | 17.12 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 12.25/ 14.66 GFLOPS | Progress: (4/20) | 6.36 s
[Task 13/25] Current/Best: 17.34/ 17.34 GFLOPS | Progress: (8/20) | 8.83 s
[Task 13/25] Current/Best: 11.10/ 20.46 GFLOPS | Progress: (12/20) | 13.07 s
[Task 13/25] Current/Best: 11.92/ 20.46 GFLOPS | Progress: (16/20) | 15.59 s
[Task 13/25] Current/Best: 1.56/ 20.46 GFLOPS | Progress: (20/20) | 19.83 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 9.71/ 11.23 GFLOPS | Progress: (4/20) | 7.59 s
[Task 14/25] Current/Best: 14.93/ 15.02 GFLOPS | Progress: (8/20) | 9.76 s
[Task 14/25] Current/Best: 10.49/ 15.02 GFLOPS | Progress: (12/20) | 13.03 s
[Task 14/25] Current/Best: 8.36/ 15.11 GFLOPS | Progress: (16/20) | 20.93 s
[Task 14/25] Current/Best: 21.24/ 21.24 GFLOPS | Progress: (20/20) | 22.86 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 9.30/ 19.90 GFLOPS | Progress: (4/20) | 7.20 s Done.
+
[Task 15/25] Current/Best: 9.58/ 19.90 GFLOPS | Progress: (8/20) | 9.90 s
[Task 15/25] Current/Best: 21.95/ 21.95 GFLOPS | Progress: (12/20) | 12.05 s
[Task 15/25] Current/Best: 15.06/ 21.95 GFLOPS | Progress: (16/20) | 13.86 s
[Task 15/25] Current/Best: 9.22/ 21.95 GFLOPS | Progress: (20/20) | 23.47 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 14.19/ 16.10 GFLOPS | Progress: (4/20) | 4.66 s
[Task 16/25] Current/Best: 12.02/ 19.12 GFLOPS | Progress: (8/20) | 6.20 s
[Task 16/25] Current/Best: 16.34/ 22.48 GFLOPS | Progress: (12/20) | 8.22 s
[Task 16/25] Current/Best: 9.99/ 22.48 GFLOPS | Progress: (16/20) | 11.36 s
[Task 16/25] Current/Best: 17.35/ 22.48 GFLOPS | Progress: (20/20) | 13.01 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 9.62/ 17.92 GFLOPS | Progress: (4/20) | 5.93 s
[Task 17/25] Current/Best: 16.38/ 17.92 GFLOPS | Progress: (8/20) | 9.88 s
[Task 17/25] Current/Best: 13.13/ 18.35 GFLOPS | Progress: (12/20) | 12.23 s
[Task 17/25] Current/Best: 3.07/ 18.35 GFLOPS | Progress: (16/20) | 15.74 s
[Task 17/25] Current/Best: 17.60/ 18.35 GFLOPS | Progress: (20/20) | 18.84 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 16.91/ 16.91 GFLOPS | Progress: (4/20) | 6.67 s
[Task 18/25] Current/Best: 11.43/ 19.34 GFLOPS | Progress: (8/20) | 9.49 s
[Task 18/25] Current/Best: 5.95/ 19.34 GFLOPS | Progress: (12/20) | 13.93 s
[Task 18/25] Current/Best: 18.35/ 19.34 GFLOPS | Progress: (16/20) | 16.10 s
[Task 18/25] Current/Best: 15.27/ 19.34 GFLOPS | Progress: (20/20) | 19.49 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 10.99/ 19.52 GFLOPS | Progress: (4/20) | 6.56 s
[Task 19/25] Current/Best: 10.38/ 19.79 GFLOPS | Progress: (8/20) | 10.24 s
[Task 19/25] Current/Best: 2.66/ 19.79 GFLOPS | Progress: (12/20) | 15.71 s
[Task 19/25] Current/Best: 1.55/ 20.33 GFLOPS | Progress: (16/20) | 19.30 s
[Task 19/25] Current/Best: 3.08/ 20.33 GFLOPS | Progress: (20/20) | 22.20 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 13.99/ 18.65 GFLOPS | Progress: (4/20) | 5.30 s
[Task 20/25] Current/Best: 16.92/ 18.65 GFLOPS | Progress: (8/20) | 7.61 s
[Task 20/25] Current/Best: 15.21/ 18.65 GFLOPS | Progress: (12/20) | 8.89 s
[Task 20/25] Current/Best: 14.43/ 20.42 GFLOPS | Progress: (16/20) | 12.95 s
[Task 20/25] Current/Best: 8.90/ 20.42 GFLOPS | Progress: (20/20) | 14.95 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 9.23/ 19.35 GFLOPS | Progress: (4/20) | 6.12 s
[Task 21/25] Current/Best: 16.43/ 19.35 GFLOPS | Progress: (8/20) | 7.69 s
[Task 21/25] Current/Best: 14.15/ 21.16 GFLOPS | Progress: (12/20) | 11.41 s Done.
Done.
-
[Task 21/25] Current/Best: 13.49/ 15.51 GFLOPS | Progress: (8/20) | 11.09 s
[Task 21/25] Current/Best: 19.22/ 19.22 GFLOPS | Progress: (12/20) | 13.48 s
[Task 21/25] Current/Best: 1.62/ 19.22 GFLOPS | Progress: (16/20) | 15.81 s
[Task 21/25] Current/Best: 12.37/ 19.22 GFLOPS | Progress: (20/20) | 19.21 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 11.20/ 18.02 GFLOPS | Progress: (4/20) | 6.62 s
[Task 22/25] Current/Best: 4.13/ 18.02 GFLOPS | Progress: (8/20) | 8.85 s
[Task 22/25] Current/Best: 13.19/ 18.02 GFLOPS | Progress: (12/20) | 14.63 s
[Task 22/25] Current/Best: 1.55/ 18.02 GFLOPS | Progress: (16/20) | 17.22 s
[Task 22/25] Current/Best: 6.67/ 19.70 GFLOPS | Progress: (20/20) | 19.92 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 11.72/ 15.98 GFLOPS | Progress: (4/20) | 7.27 s
[Task 23/25] Current/Best: 6.11/ 23.08 GFLOPS | Progress: (8/20) | 9.82 s
[Task 23/25] Current/Best: 11.45/ 23.08 GFLOPS | Progress: (12/20) | 13.28 s
[Task 23/25] Current/Best: 18.39/ 23.08 GFLOPS | Progress: (16/20) | 17.24 s
[Task 23/25] Current/Best: 11.55/ 23.08 GFLOPS | Progress: (20/20) | 20.23 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 1.84/ 6.88 GFLOPS | Progress: (4/20) | 13.89 s
[Task 24/25] Current/Best: 1.49/ 6.88 GFLOPS | Progress: (8/20) | 26.54 s
[Task 24/25] Current/Best: 0.49/ 6.88 GFLOPS | Progress: (12/20) | 34.21 s
[Task 24/25] Current/Best: 5.62/ 6.88 GFLOPS | Progress: (16/20) | 45.26 s
[Task 24/25] Current/Best: 1.94/ 6.88 GFLOPS | Progress: (20/20) | 56.32 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
- Done.
-
[Task 25/25] Current/Best: 3.84/ 4.27 GFLOPS | Progress: (4/20) | 15.76 s
[Task 25/25] Current/Best: 4.49/ 4.76 GFLOPS | Progress: (8/20) | 28.56 s
[Task 25/25] Current/Best: 5.74/ 5.74 GFLOPS | Progress: (12/20) | 39.64 s
[Task 25/25] Current/Best: 5.20/ 5.74 GFLOPS | Progress: (16/20) | 50.72 s
[Task 25/25] Current/Best: 2.67/ 5.74 GFLOPS | Progress: (20/20) | 61.77 s
+
[Task 21/25] Current/Best: 9.56/ 21.16 GFLOPS | Progress: (16/20) | 14.13 s
[Task 21/25] Current/Best: 6.29/ 21.16 GFLOPS | Progress: (20/20) | 17.20 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 15.58/ 15.58 GFLOPS | Progress: (4/20) | 5.34 s
[Task 22/25] Current/Best: 14.35/ 15.58 GFLOPS | Progress: (8/20) | 7.97 s
[Task 22/25] Current/Best: 10.63/ 18.45 GFLOPS | Progress: (12/20) | 10.04 s
[Task 22/25] Current/Best: 10.40/ 18.45 GFLOPS | Progress: (16/20) | 12.97 s
[Task 22/25] Current/Best: 16.47/ 20.31 GFLOPS | Progress: (20/20) | 14.59 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 20.03/ 20.03 GFLOPS | Progress: (4/20) | 5.19 s
[Task 23/25] Current/Best: 18.31/ 20.03 GFLOPS | Progress: (8/20) | 8.05 s
[Task 23/25] Current/Best: 10.63/ 20.03 GFLOPS | Progress: (12/20) | 12.88 s
[Task 23/25] Current/Best: 18.71/ 20.03 GFLOPS | Progress: (16/20) | 15.16 s
[Task 23/25] Current/Best: 11.48/ 20.03 GFLOPS | Progress: (20/20) | 18.82 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 6.27/ 6.27 GFLOPS | Progress: (4/20) | 11.20 s
[Task 24/25] Current/Best: 3.45/ 6.93 GFLOPS | Progress: (8/20) | 22.18 s
[Task 24/25] Current/Best: 1.54/ 6.93 GFLOPS | Progress: (12/20) | 33.18 s
[Task 24/25] Current/Best: 2.83/ 6.93 GFLOPS | Progress: (16/20) | 45.42 s
[Task 24/25] Current/Best: 7.03/ 7.03 GFLOPS | Progress: (20/20) | 53.79 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
[Task 25/25] Current/Best: 1.55/ 5.76 GFLOPS | Progress: (4/20) | 5.70 s
[Task 25/25] Current/Best: 9.17/ 9.17 GFLOPS | Progress: (8/20) | 16.08 s
[Task 25/25] Current/Best: 7.32/ 9.17 GFLOPS | Progress: (12/20) | 29.10 s
[Task 25/25] Current/Best: 1.54/ 9.17 GFLOPS | Progress: (16/20) | 40.13 s
[Task 25/25] Current/Best: 6.93/ 9.17 GFLOPS | Progress: (20/20) | 46.66 s
@@ -702,7 +701,7 @@ Verify that the optimized model runs and produces the same results:
.. code-block:: none
- class='n02123045 tabby, tabby cat' with probability=0.621104
+ class='n02123045 tabby, tabby cat' with probability=0.621105
class='n02123159 tiger cat' with probability=0.356378
class='n02124075 Egyptian cat' with probability=0.019712
class='n02129604 tiger, Panthera tigris' with probability=0.001215
@@ -760,8 +759,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 430.4348260200004, 'median': 430.0954028000092, 'std': 2.5709889563337005}
- unoptimized: {'mean': 528.2606210899996, 'median': 527.0514347999949, 'std': 4.135171668509399}
+ optimized: {'mean': 415.6101647899777, 'median': 414.97993934999613, 'std': 2.3320012765354075}
+ unoptimized: {'mean': 516.5330643200014, 'median': 516.7005607999727, 'std': 2.309861836840107}
@@ -784,7 +783,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 13 minutes 43.274 seconds)
+ **Total running time of the script:** ( 12 minutes 54.598 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 83138ebd3e..ea258591cf 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -274,7 +274,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.28e-07 secs/op
+ 1.43e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 84535a6ea3..8825fd1d21 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -270,7 +270,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0x148499f0)), stage(b, placeholder(b, 0x27940570)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T [...]
+ [stage(a, placeholder(a, 0x123148d0)), stage(b, placeholder(b, 0x1233fe40)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index ddf4ee8b57..702fdc6e22 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,24 +5,24 @@
Computation times
=================
-**17:36.980** total execution time for **tutorial** files:
+**16:39.383** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 13:43.274 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 12:54.598 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:39.679 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:34.545 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:06.505 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:01.944 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:40.485 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:38.407 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:24.090 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:28.097 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.804 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.882 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.934 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.732 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.208 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.178 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.000 | 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 3d9d04471f..e9d6d32dfa 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -285,7 +285,7 @@ helper function to run a profile of the TVM generated code.
.. code-block:: none
- Numpy running time: 0.000008
+ Numpy running time: 0.000007
naive: 0.000007
@@ -444,7 +444,7 @@ factor to be the number of threads on your CPU.
.. code-block:: none
- vector: 0.000040
+ vector: 0.000041
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -498,10 +498,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 7.936710001104074e-06 1.0
- naive 6.878700000000001e-06 0.8666941338467836
- parallel 6.9646e-06 0.8775172582885297
- vector 4.0432800000000004e-05 5.094403095788483
+ numpy 7.4954699994123074e-06 1.0
+ naive 6.6504e-06 0.8872559026347159
+ parallel 6.9207000000000005e-06 0.9233176839534583
+ vector 4.06571e-05 5.424222897721927
@@ -922,7 +922,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.019826
+ Numpy running time: 0.019090
@@ -980,7 +980,7 @@ optimizations.
.. code-block:: none
- none: 3.693602
+ none: 3.451939
@@ -1080,7 +1080,7 @@ schedule.
.. code-block:: none
- blocking: 0.349235
+ blocking: 0.312328
@@ -1164,7 +1164,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.374328
+ vectorization: 0.348143
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1230,7 +1230,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.149761
+ loop permutation: 0.118454
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1321,7 +1321,7 @@ optimized schedule.
.. code-block:: none
- array packing: 0.109916
+ array packing: 0.109026
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1404,7 +1404,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.114692
+ block caching: 0.111010
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1478,7 +1478,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.149886
+ parallelization: 0.146741
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1548,13 +1548,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.6936021151 1.0
- blocking 0.34923484650000003 0.0945512904793604
- vectorization 0.37432799730000005 0.10134497047467322
- loop permutation 0.14976123270000002 0.04054611948800701
- array packing 0.1099163436 0.029758577176097412
- block caching 0.1146916379 0.031051432808943704
- parallelization 0.1498861889 0.04057994993213881
+ none 3.4519386253 1.0
+ blocking 0.3123282837 0.0904790952570474
+ vectorization 0.34814288050000003 0.1008543077644504
+ loop permutation 0.1184544058 0.034315327894830515
+ array packing 0.10902590059999999 0.03158396264664897
+ block caching 0.11101049140000001 0.032158883297165326
+ parallelization 0.14674148110000002 0.04250987547243748
@@ -1596,7 +1596,7 @@ the computation for specific platforms.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 6.505 seconds)
+ **Total running time of the script:** ( 1 minutes 1.944 seconds)
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index bb89e50f12..7d3ef9527c 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-706a5b2dd5ab8f22a88b84eeb4fd94f318f1e9a5
+eca6edf7faf8d8ec0a6ea09b5d6f1ba06153a594
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index 08d30cb2f1..ac78e66950 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -590,7 +590,7 @@ 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 28.330 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 24.945 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_keras.html b/docs/how_to/compile_models/from_keras.html
index edd3732e34..c87458d826 100644
--- a/docs/how_to/compile_models/from_keras.html
+++ b/docs/how_to/compile_models/from_keras.html
@@ -511,7 +511,7 @@ Tensorflow is also required since it’s used as the default backend of keras.</
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Relay top-1 id: 285, class name: Egyptian cat
1/1 [==============================] - ETA: 0s
-1/1 [==============================] - 2s 2s/step
+1/1 [==============================] - 1s 1s/step
Keras top-1 id: 285, class name: Egyptian cat
</pre></div>
</div>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index a77b80d612..d4aab9bc38 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -444,7 +444,7 @@
<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.zip3bdf3f20-1bc9-4501-aa3a-7b59d9a0477e 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.zip1efd7c6f-5a68-4c9c-96e2-111990adc926 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 a65d4b5d3d..fdaceffdee 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -454,12 +454,14 @@ Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdo
<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, 52.9MB/s]
- 39%|###8 | 16.0M/41.5M [00:00<00:00, 52.3MB/s]
- 58%|#####7 | 24.0M/41.5M [00:00<00:00, 58.6MB/s]
- 77%|#######7 | 32.0M/41.5M [00:00<00:00, 57.9MB/s]
- 91%|######### | 37.6M/41.5M [00:00<00:00, 49.2MB/s]
-100%|##########| 41.5M/41.5M [00:01<00:00, 38.0MB/s]
+ 15%|#5 | 6.33M/41.5M [00:00<00:00, 48.1MB/s]
+ 26%|##6 | 10.9M/41.5M [00:00<00:00, 44.8MB/s]
+ 37%|###6 | 15.2M/41.5M [00:00<00:00, 34.3MB/s]
+ 45%|####4 | 18.6M/41.5M [00:00<00:00, 33.5MB/s]
+ 63%|######2 | 26.1M/41.5M [00:00<00:00, 46.9MB/s]
+ 77%|#######7 | 32.0M/41.5M [00:00<00:00, 46.4MB/s]
+ 92%|#########2| 38.3M/41.5M [00:00<00:00, 41.8MB/s]
+100%|##########| 41.5M/41.5M [00:01<00:00, 41.1MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_paddle.html b/docs/how_to/compile_models/from_paddle.html
index fd0fe9505f..cd86dc8824 100644
--- a/docs/how_to/compile_models/from_paddle.html
+++ b/docs/how_to/compile_models/from_paddle.html
@@ -489,7 +489,6 @@ To begin, we’ll install PaddlePaddle>=2.1.3:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>TVM prediction top-1 id: 282, class name: 282: 'tiger cat',
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 4.701 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-paddle-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/16269b77359771348d507395692524cf/from_paddle.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_paddle.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index 3f02d3ecce..81b5df1ab0 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -437,13 +437,13 @@ be unstable.</p>
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]
- 14%|#4 | 6.30M/44.7M [00:00<00:00, 56.2MB/s]
- 26%|##6 | 11.7M/44.7M [00:00<00:00, 42.4MB/s]
- 36%|###5 | 16.0M/44.7M [00:00<00:00, 40.6MB/s]
- 58%|#####8 | 26.0M/44.7M [00:00<00:00, 61.8MB/s]
- 79%|#######9 | 35.4M/44.7M [00:00<00:00, 73.6MB/s]
- 99%|#########8| 44.2M/44.7M [00:00<00:00, 78.9MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 66.8MB/s]
+ 18%|#7 | 7.99M/44.7M [00:00<00:00, 59.8MB/s]
+ 36%|###5 | 16.0M/44.7M [00:00<00:00, 57.5MB/s]
+ 54%|#####3 | 24.0M/44.7M [00:00<00:00, 63.9MB/s]
+ 68%|######7 | 30.2M/44.7M [00:00<00:00, 57.1MB/s]
+ 80%|#######9 | 35.7M/44.7M [00:00<00:00, 48.9MB/s]
+ 91%|######### | 40.5M/44.7M [00:00<00:00, 45.9MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 55.5MB/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 299a2bfe6b..f8cf11aa31 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -657,7 +657,7 @@ 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 40.711 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 31.607 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 2c5c531341..0c35e303a8 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<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>07:35.835</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>07:03.903</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -354,43 +354,43 @@
</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>01:40.711</p></td>
+<td><p>01:31.607</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>01:28.330</p></td>
+<td><p>01:24.945</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>01:04.701</p></td>
+<td><p>00:59.391</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>00:41.806</p></td>
+<td><p>00:38.869</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>00:37.252</p></td>
+<td><p>00:34.177</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>00:33.273</p></td>
+<td><p>00:31.393</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>00:30.294</p></td>
+<td><p>00:29.298</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>00:29.209</p></td>
+<td><p>00:26.631</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>00:27.269</p></td>
+<td><p>00:24.751</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>00:02.991</p></td>
+<td><p>00:02.841</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/deploy_models/deploy_model_on_adreno.html b/docs/how_to/deploy_models/deploy_model_on_adreno.html
index 92070f813f..924d92375c 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno.html
@@ -835,10 +835,10 @@ Top5 predictions:
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 3354.0283 3351.9853 3365.8622 3349.3537 4.8277
+ 3345.4809 3344.7137 3354.2131 3340.3818 3.6215
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 8.105 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 5.981 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-model-on-adreno-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/2387d8448da213eb625e6b3d916327d4/deploy_model_on_adreno.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_model_on_adreno.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html b/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
index 67007baa19..d1f78aeb95 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
@@ -443,32 +443,29 @@ to run this tutorial with a real device over rpc.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5
8192/102967424 [..............................] - ETA: 0s
- 6635520/102967424 [>.............................] - ETA: 0s
- 8380416/102967424 [=>............................] - ETA: 1s
- 15024128/102967424 [===>..........................] - ETA: 1s
- 16769024/102967424 [===>..........................] - ETA: 1s
- 23412736/102967424 [=====>........................] - ETA: 1s
+ 8380416/102967424 [=>............................] - ETA: 0s
+ 16769024/102967424 [===>..........................] - ETA: 0s
+ 19562496/102967424 [====>.........................] - ETA: 0s
25157632/102967424 [======>.......................] - ETA: 1s
+ 27992064/102967424 [=======>......................] - ETA: 1s
33546240/102967424 [========>.....................] - ETA: 1s
- 41934848/102967424 [===========>..................] - ETA: 1s
- 48578560/102967424 [=============>................] - ETA: 1s
- 50323456/102967424 [=============>................] - ETA: 1s
- 56967168/102967424 [===============>..............] - ETA: 0s
- 58712064/102967424 [================>.............] - ETA: 1s
- 65355776/102967424 [==================>...........] - ETA: 0s
+ 35741696/102967424 [=========>....................] - ETA: 1s
+ 41934848/102967424 [===========>..................] - ETA: 0s
+ 42065920/102967424 [===========>..................] - ETA: 1s
+ 50323456/102967424 [=============>................] - ETA: 0s
+ 58712064/102967424 [================>.............] - ETA: 0s
+ 60932096/102967424 [================>.............] - ETA: 0s
67100672/102967424 [==================>...........] - ETA: 0s
69296128/102967424 [===================>..........] - ETA: 0s
75202560/102967424 [====================>.........] - ETA: 0s
- 75489280/102967424 [====================>.........] - ETA: 0s
- 76972032/102967424 [=====================>........] - ETA: 0s
+ 76529664/102967424 [=====================>........] - ETA: 0s
82124800/102967424 [======================>.......] - ETA: 0s
- 83877888/102967424 [=======================>......] - ETA: 0s
+ 84926464/102967424 [=======================>......] - ETA: 0s
90521600/102967424 [=========================>....] - ETA: 0s
92266496/102967424 [=========================>....] - ETA: 0s
- 98910208/102967424 [===========================>..] - ETA: 0s
-100646912/102967424 [============================>.] - ETA: 0s
+ 94461952/102967424 [==========================>...] - ETA: 0s
102850560/102967424 [============================>.] - ETA: 0s
-102967424/102967424 [==============================] - 3s 0us/step
+102967424/102967424 [==============================] - 2s 0us/step
</pre></div>
</div>
</div>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 2cf85cd7d5..a4f019ec83 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -667,7 +667,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 17.2645 17.0776 17.8742 16.7083 0.4020
+ 16.2511 16.0867 17.7078 15.9810 0.4937
</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 ae85a6a0c3..7eb28bc47d 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -459,35 +459,33 @@ be unstable.</p>
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]
- 5%|4 | 7.99M/170M [00:00<00:02, 56.6MB/s]
- 9%|8 | 14.5M/170M [00:00<00:02, 62.5MB/s]
- 12%|#2 | 20.6M/170M [00:00<00:02, 63.0MB/s]
- 16%|#5 | 26.7M/170M [00:00<00:04, 32.7MB/s]
- 19%|#8 | 32.0M/170M [00:00<00:04, 30.9MB/s]
- 24%|##3 | 40.0M/170M [00:01<00:03, 39.1MB/s]
- 28%|##8 | 48.0M/170M [00:01<00:03, 42.4MB/s]
- 33%|###2 | 56.0M/170M [00:01<00:02, 48.7MB/s]
- 38%|###7 | 64.0M/170M [00:01<00:01, 55.7MB/s]
- 42%|####2 | 72.0M/170M [00:01<00:01, 52.4MB/s]
- 46%|####6 | 78.3M/170M [00:01<00:02, 46.4MB/s]
- 49%|####8 | 83.1M/170M [00:02<00:02, 38.2MB/s]
- 52%|#####1 | 88.0M/170M [00:02<00:02, 36.8MB/s]
- 55%|#####4 | 93.2M/170M [00:02<00:02, 38.2MB/s]
- 57%|#####7 | 97.0M/170M [00:02<00:02, 35.9MB/s]
- 61%|######1 | 104M/170M [00:02<00:01, 43.3MB/s]
- 64%|######3 | 108M/170M [00:02<00:01, 42.8MB/s]
- 66%|######6 | 112M/170M [00:02<00:01, 40.1MB/s]
- 71%|####### | 120M/170M [00:02<00:01, 44.5MB/s]
- 74%|#######4 | 126M/170M [00:03<00:01, 45.2MB/s]
- 79%|#######9 | 134M/170M [00:03<00:00, 47.6MB/s]
- 82%|########1 | 139M/170M [00:03<00:00, 47.4MB/s]
- 85%|########4 | 144M/170M [00:03<00:00, 43.1MB/s]
- 88%|########8 | 150M/170M [00:03<00:00, 44.2MB/s]
- 91%|#########1| 155M/170M [00:03<00:00, 39.0MB/s]
- 93%|#########3| 158M/170M [00:04<00:00, 29.5MB/s]
- 95%|#########5| 161M/170M [00:04<00:00, 26.4MB/s]
- 98%|#########7| 166M/170M [00:04<00:00, 29.8MB/s]
-100%|##########| 170M/170M [00:04<00:00, 40.3MB/s]
+ 5%|4 | 7.99M/170M [00:00<00:03, 42.9MB/s]
+ 8%|8 | 14.3M/170M [00:00<00:03, 48.6MB/s]
+ 11%|#1 | 19.1M/170M [00:00<00:03, 49.0MB/s]
+ 14%|#4 | 23.9M/170M [00:00<00:03, 39.1MB/s]
+ 16%|#6 | 27.8M/170M [00:00<00:04, 36.8MB/s]
+ 19%|#8 | 31.8M/170M [00:00<00:03, 38.2MB/s]
+ 21%|## | 35.6M/170M [00:00<00:03, 35.2MB/s]
+ 24%|##3 | 40.0M/170M [00:01<00:03, 36.6MB/s]
+ 28%|##8 | 48.0M/170M [00:01<00:03, 39.2MB/s]
+ 33%|###2 | 56.0M/170M [00:01<00:02, 43.4MB/s]
+ 38%|###7 | 64.0M/170M [00:01<00:02, 49.0MB/s]
+ 42%|####2 | 72.0M/170M [00:01<00:01, 52.2MB/s]
+ 47%|####7 | 80.0M/170M [00:01<00:01, 53.1MB/s]
+ 52%|#####1 | 88.0M/170M [00:02<00:01, 49.5MB/s]
+ 56%|#####5 | 94.7M/170M [00:02<00:01, 53.8MB/s]
+ 60%|#####9 | 102M/170M [00:02<00:01, 58.6MB/s]
+ 63%|######3 | 108M/170M [00:02<00:01, 52.2MB/s]
+ 66%|######6 | 113M/170M [00:02<00:01, 50.4MB/s]
+ 71%|####### | 120M/170M [00:02<00:00, 57.6MB/s]
+ 75%|#######5 | 128M/170M [00:02<00:00, 56.6MB/s]
+ 79%|#######9 | 134M/170M [00:02<00:00, 55.1MB/s]
+ 82%|########2 | 140M/170M [00:03<00:00, 52.1MB/s]
+ 85%|########5 | 145M/170M [00:03<00:00, 42.2MB/s]
+ 89%|########9 | 152M/170M [00:03<00:00, 42.5MB/s]
+ 94%|#########4| 160M/170M [00:03<00:00, 44.2MB/s]
+ 99%|#########8| 168M/170M [00:03<00:00, 51.8MB/s]
+100%|##########| 170M/170M [00:03<00:00, 48.3MB/s]
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: 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)
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: 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=& [...]
@@ -585,7 +583,7 @@ 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> ( 4 minutes 12.708 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 48.442 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 52edce790f..bd0011ec46 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -500,9 +500,8 @@ training. Other models require a full post training calibration.</p>
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]
- 47%|####6 | 6.30M/13.6M [00:00<00:00, 47.7MB/s]
- 80%|######## | 10.9M/13.6M [00:00<00:00, 42.5MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 41.0MB/s]
+ 59%|#####8 | 7.99M/13.6M [00:00<00:00, 55.3MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 80.8MB/s]
</pre></div>
</div>
</div>
@@ -593,7 +592,7 @@ 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)
- 91.1190 91.0954 91.8840 90.7369 0.2098
+ 90.4576 90.4198 93.3698 90.1384 0.3345
</pre></div>
</div>
<div class="admonition note">
@@ -632,7 +631,7 @@ 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 27.939 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 20.934 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 9d8f3b85b3..cf70732684 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -585,7 +585,7 @@ 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)
- 123.4716 123.4034 127.8491 121.4705 0.8379
+ 121.4166 121.2992 125.6682 120.6915 0.6146
</pre></div>
</div>
<div class="admonition note">
@@ -613,7 +613,7 @@ 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> ( 2 minutes 39.446 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 35.613 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 f62abf8adc..ff62bb7455 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -526,7 +526,7 @@ 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 58.131 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 4.876 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 dbf9d0b3fd..d6b781106d 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -468,24 +468,22 @@ 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 | 6117/132723 [00:00<00:02, 61159.41KB/s]
- 11%|# | 14375/132723 [00:00<00:01, 73756.01KB/s]
- 16%|#6 | 21751/132723 [00:00<00:01, 73587.55KB/s]
- 23%|##2 | 29905/132723 [00:00<00:01, 76719.15KB/s]
- 28%|##8 | 37578/132723 [00:00<00:01, 54915.12KB/s]
- 35%|###4 | 45831/132723 [00:00<00:01, 62348.13KB/s]
- 40%|###9 | 52744/132723 [00:00<00:01, 60531.76KB/s]
- 46%|####5 | 60891/132723 [00:00<00:01, 66227.03KB/s]
- 51%|#####1 | 67919/132723 [00:01<00:01, 63708.72KB/s]
- 57%|#####7 | 76152/132723 [00:01<00:00, 68816.46KB/s]
- 63%|######2 | 83294/132723 [00:01<00:00, 58507.35KB/s]
- 69%|######8 | 91409/132723 [00:01<00:00, 64230.02KB/s]
- 74%|#######4 | 98288/132723 [00:01<00:00, 51991.29KB/s]
- 78%|#######8 | 104103/132723 [00:01<00:00, 53071.67KB/s]
- 85%|########4 | 112282/132723 [00:01<00:00, 60168.35KB/s]
- 89%|########9 | 118775/132723 [00:01<00:00, 60947.09KB/s]
- 96%|#########5| 126989/132723 [00:02<00:00, 66670.43KB/s]
-100%|##########| 132723/132723 [00:02<00:00, 63348.09KB/s]
+ 5%|5 | 7061/132723 [00:00<00:01, 70603.00KB/s]
+ 12%|#1 | 15712/132723 [00:00<00:01, 79955.80KB/s]
+ 18%|#7 | 23708/132723 [00:00<00:01, 59640.98KB/s]
+ 24%|##4 | 32324/132723 [00:00<00:01, 68318.18KB/s]
+ 31%|### | 40863/132723 [00:00<00:01, 73754.54KB/s]
+ 37%|###7 | 49377/132723 [00:00<00:01, 77314.50KB/s]
+ 44%|####3 | 57952/132723 [00:00<00:00, 79909.66KB/s]
+ 50%|##### | 66587/132723 [00:00<00:00, 81878.30KB/s]
+ 57%|#####6 | 75196/132723 [00:00<00:00, 83156.29KB/s]
+ 63%|######2 | 83615/132723 [00:01<00:00, 77912.83KB/s]
+ 69%|######9 | 92197/132723 [00:01<00:00, 80185.58KB/s]
+ 76%|#######5 | 100789/132723 [00:01<00:00, 81848.19KB/s]
+ 82%|########2 | 109426/132723 [00:01<00:00, 83175.37KB/s]
+ 89%|########8 | 118025/132723 [00:01<00:00, 84005.95KB/s]
+ 95%|#########5| 126667/132723 [00:01<00:00, 84720.83KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 79461.78KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -524,7 +522,7 @@ 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> ( 4 minutes 24.382 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> ( 4 minutes 8.435 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 360f384747..ca648e7bb3 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<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>18:41.806</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>17:42.479</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -354,47 +354,47 @@
</colgroup>
<tbody>
<tr class="row-odd"><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>04:24.382</p></td>
+<td><p>04:08.435</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><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>04:12.708</p></td>
+<td><p>03:48.442</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>02:39.446</p></td>
+<td><p>02:35.613</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>01:58.131</p></td>
+<td><p>02:04.876</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>01:27.939</p></td>
+<td><p>01:20.934</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_adreno.html#sphx-glr-how-to-deploy-models-deploy-model-on-adreno-py"><span class="std std-ref">Deploy the Pretrained Model on Adreno™</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_adreno.py</span></code>)</p></td>
-<td><p>01:08.105</p></td>
+<td><p>01:05.981</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_adreno_tvmc.html#sphx-glr-how-to-deploy-models-deploy-model-on-adreno-tvmc-py"><span class="std std-ref">Deploy the Pretrained Model on Adreno™ with tvmc Interface</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_adreno_tvmc.py</span></code>)</p></td>
-<td><p>00:59.271</p></td>
+<td><p>00:54.553</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>00:49.280</p></td>
+<td><p>00:45.143</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_nano.html#sphx-glr-how-to-deploy-models-deploy-model-on-nano-py"><span class="std std-ref">Deploy the Pretrained Model on Jetson Nano</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_nano.py</span></code>)</p></td>
-<td><p>00:31.685</p></td>
+<td><p>00:29.529</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><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>00:30.852</p></td>
+<td><p>00:28.968</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><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>
-<td><p>00:00.007</p></td>
+<td><p>00:00.006</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
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 de5cf6db3e..a85dc91540 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -624,7 +624,7 @@ 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.zipa3a78f71-d8cf-4767-9bb9-3baef1043d1c 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.zip4abbc63d-ab71-479f-9533-9b14d243364a 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 96a6b6855f..28f71caf5a 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<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>01:02.783</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:57.963</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,19 +354,19 @@
</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>00:58.392</p></td>
+<td><p>00:53.923</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>00:03.149</p></td>
+<td><p>00:02.898</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>00:01.233</p></td>
+<td><p>00:01.135</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>00:00.008</p></td>
+<td><p>00:00.007</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 c3e94c3263..e1dcc3dca9 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -531,10 +531,10 @@ 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: 24496us [24496us] (48.57%; 48.57%)
-FoldScaleAxis: 25940us [11us] (51.43%; 51.43%)
- FoldConstant: 25930us [1903us] (51.41%; 99.96%)
- InferType: 24027us [24027us] (47.64%; 92.66%)
+InferType: 23880us [23880us] (49.04%; 49.04%)
+FoldScaleAxis: 24818us [9us] (50.96%; 50.96%)
+ FoldConstant: 24808us [1785us] (50.94%; 99.96%)
+ InferType: 23024us [23024us] (47.28%; 92.81%)
</pre></div>
</div>
</div>
@@ -556,10 +556,10 @@ 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: 24332us [24332us] (48.34%; 48.34%)
-FoldScaleAxis: 26006us [10us] (51.66%; 51.66%)
- FoldConstant: 25996us [2073us] (51.64%; 99.96%)
- InferType: 23923us [23923us] (47.53%; 92.03%)
+InferType: 23006us [23006us] (48.04%; 48.04%)
+FoldScaleAxis: 24886us [8us] (51.96%; 51.96%)
+ FoldConstant: 24878us [1785us] (51.95%; 99.97%)
+ InferType: 23093us [23093us] (48.22%; 92.82%)
</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 dbf3145279..0650b91a87 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -580,7 +580,7 @@ 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: 53.541183 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 40.979232 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 ac8d9f31be..bafddcc480 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -862,7 +862,7 @@ 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: 12.221783 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 11.890704 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 bf3f910b69..61531e69cb 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -477,8 +477,8 @@ 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: 0.020033
-Baseline: 3.752375
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018918
+Baseline: 3.430278
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -537,7 +537,7 @@ 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: 0.348482
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.308531
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -594,7 +594,7 @@ 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: 0.369061
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.342998
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -649,7 +649,7 @@ 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: 0.152658
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.117401
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -726,7 +726,7 @@ 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: 0.114943
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.108893
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -804,7 +804,7 @@ 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: 0.116410
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.110965
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -884,7 +884,7 @@ class Module:
<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: 0.152030
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.147871
</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 dcf5293cbe..a41d2fc4ea 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -345,7 +345,7 @@
<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>00:38.659</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.911</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,15 +354,15 @@
</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>00:35.436</p></td>
+<td><p>00:32.735</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>00:01.984</p></td>
+<td><p>00:01.995</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>00:01.240</p></td>
+<td><p>00:01.181</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 b25d292370..1c72e1618e 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -345,7 +345,7 @@
<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>11:15.168</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>10:59.004</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -354,27 +354,27 @@
</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>06:57.183</p></td>
+<td><p>06:51.242</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>01:53.239</p></td>
+<td><p>01:47.856</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>01:18.112</p></td>
+<td><p>01:14.798</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>00:35.326</p></td>
+<td><p>00:35.488</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><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>00:15.980</p></td>
+<td><p>00:15.206</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><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>00:15.327</p></td>
+<td><p>00:14.413</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 e37d59bccc..f0128d5ca1 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
@@ -494,9 +494,6 @@ file and apply it.</p>
<span class="k">del</span> <span class="n">measure_ctx</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>.T
-</pre></div>
-</div>
<p>We can lower the schedule to see the IR after auto-scheduling.
The auto-scheduler correctly performs optimizations including multi-level tiling,
cooperative fetching, unrolling and operator fusion.</p>
@@ -513,479 +510,577 @@ class Module:
@T.prim_func
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
- blockIdx_x = T.launch_thread("blockIdx.x", 28)
- conv2d_nchw = T.allocate([14], "float32", "local")
- pad_temp_shared = T.allocate([72], "float32", "shared")
- kernel_shared = T.allocate([3072], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 64)
- conv2d_nchw_1 = T.Buffer((14,), data=conv2d_nchw, scope="local", align=32)
+ blockIdx_x = T.launch_thread("blockIdx.x", 16)
+ conv2d_nchw = T.allocate([16], "float32", "local")
+ pad_temp_shared = T.allocate([2592], "float32", "shared")
+ kernel_shared = T.allocate([9216], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 98)
+ conv2d_nchw_1 = T.Buffer((64,), data=conv2d_nchw, scope="local", align=32)
conv2d_nchw_1[0] = T.float32(0)
+ conv2d_nchw_1[8] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
+ conv2d_nchw_1[9] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
+ conv2d_nchw_1[10] = T.float32(0)
conv2d_nchw_1[3] = T.float32(0)
+ conv2d_nchw_1[11] = T.float32(0)
conv2d_nchw_1[4] = T.float32(0)
+ conv2d_nchw_1[12] = T.float32(0)
conv2d_nchw_1[5] = T.float32(0)
+ conv2d_nchw_1[13] = T.float32(0)
conv2d_nchw_1[6] = T.float32(0)
+ conv2d_nchw_1[14] = T.float32(0)
conv2d_nchw_1[7] = T.float32(0)
- conv2d_nchw_1[8] = T.float32(0)
- conv2d_nchw_1[9] = T.float32(0)
- conv2d_nchw_1[10] = T.float32(0)
- conv2d_nchw_1[11] = T.float32(0)
- conv2d_nchw_1[12] = T.float32(0)
- conv2d_nchw_1[13] = T.float32(0)
- for rc_outer_outer, ry_outer_outer in T.grid(64, 3):
- cse_var_2: T.int32 = rc_outer_outer * 72
- cse_var_1: T.int32 = ry_outer_outer * 3
- pad_temp_shared_1 = T.Buffer((72,), data=pad_temp_shared, scope="shared")
- with T.launch_thread("threadIdx.x", 64) as threadIdx_x_1:
- data_1 = T.Buffer((25088,), data=data.data)
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= threadIdx_x_1 * 4 % 9 and threadIdx_x_1 * 4 % 9 < 8, data_1[rc_outer_outer * 392 + threadIdx_x_1 * 4 // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + threadIdx_x_1 * 4 % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 1] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 1) % 9 and (threadIdx_x_1 * 4 + 1) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 1) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 1) % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 2] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 2) % 9 and (threadIdx_x_1 * 4 + 2) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 2) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 2) % 9 - 8], T.float32(0))
- if T.likely(threadIdx_x_1 < 18):
- pad_temp_shared_1[threadIdx_x_1 * 4 + 3] = T.if_then_else(1 <= ry_outer_outer + blockIdx_x % 7 and ry_outer_outer + blockIdx_x % 7 < 8 and 1 <= (threadIdx_x_1 * 4 + 3) % 9 and (threadIdx_x_1 * 4 + 3) % 9 < 8, data_1[rc_outer_outer * 392 + (threadIdx_x_1 * 4 + 3) // 9 * 49 + ry_outer_outer * 7 + blockIdx_x % 7 * 7 + (threadIdx_x_1 * 4 + 3) % 9 - 8], T.float32(0))
+ conv2d_nchw_1[15] = T.float32(0)
+ for rc_outer_outer in range(16):
+ cse_var_2: T.int32 = rc_outer_outer * 1568
+ cse_var_1: T.int32 = rc_outer_outer * 288
threadIdx_x_1 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((3072,), data=kernel_shared, scope="shared")
+ pad_temp_shared_1 = T.Buffer((2592,), data=pad_temp_shared, scope="shared")
+ data_1 = T.Buffer((25088,), data=data.data)
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(9 <= threadIdx_x_1 % 81 and threadIdx_x_1 % 81 < 72 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 // 81 * 49 + threadIdx_x_1 % 81 // 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 98] = T.if_then_else(9 <= (threadIdx_x_1 + 17) % 81 and (threadIdx_x_1 + 17) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 98) // 81 * 49 + (threadIdx_x_1 + 17) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 196] = T.if_then_else(9 <= (threadIdx_x_1 + 34) % 81 and (threadIdx_x_1 + 34) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 196) // 81 * 49 + (threadIdx_x_1 + 34) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 294] = T.if_then_else(9 <= (threadIdx_x_1 + 51) % 81 and (threadIdx_x_1 + 51) % 81 < 72 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 294) // 81 * 49 + (threadIdx_x_1 + 51) % 81 // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 392] = T.if_then_else(9 <= (threadIdx_x_1 + 68) % 81 and (threadIdx_x_1 + 68) % 81 < 72 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 392) // 81 * 49 + (threadIdx_x_1 + 68) % 81 // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 490] = T.if_then_else(9 <= (threadIdx_x_1 + 4) % 81 and (threadIdx_x_1 + 4) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 490) // 81 * 49 + (threadIdx_x_1 + 4) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 588] = T.if_then_else(9 <= (threadIdx_x_1 + 21) % 81 and (threadIdx_x_1 + 21) % 81 < 72 and 1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 588) // 81 * 49 + (threadIdx_x_1 + 21) % 81 // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 686] = T.if_then_else(9 <= (threadIdx_x_1 + 38) % 81 and (threadIdx_x_1 + 38) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 686) // 81 * 49 + (threadIdx_x_1 + 38) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 784] = T.if_then_else(9 <= (threadIdx_x_1 + 55) % 81 and (threadIdx_x_1 + 55) % 81 < 72 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 784) // 81 * 49 + (threadIdx_x_1 + 55) % 81 // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 882] = T.if_then_else(1 <= (threadIdx_x_1 // 9 + 8) % 9 and (threadIdx_x_1 + 72) % 81 < 72 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 882) // 81 * 49 + (threadIdx_x_1 // 9 + 8) % 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 980] = T.if_then_else(9 <= (threadIdx_x_1 + 8) % 81 and (threadIdx_x_1 + 8) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 980) // 81 * 49 + (threadIdx_x_1 + 8) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1078] = T.if_then_else(9 <= (threadIdx_x_1 + 25) % 81 and (threadIdx_x_1 + 25) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1078) // 81 * 49 + (threadIdx_x_1 + 25) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1176] = T.if_then_else(9 <= (threadIdx_x_1 + 42) % 81 and (threadIdx_x_1 + 42) % 81 < 72 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1176) // 81 * 49 + (threadIdx_x_1 + 42) % 81 // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1274] = T.if_then_else(9 <= (threadIdx_x_1 + 59) % 81 and (threadIdx_x_1 + 59) % 81 < 72 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1274) // 81 * 49 + (threadIdx_x_1 + 59) % 81 // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1372] = T.if_then_else(9 <= (threadIdx_x_1 + 76) % 81 and (threadIdx_x_1 + 76) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1372) // 81 * 49 + (threadIdx_x_1 + 76) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1470] = T.if_then_else(9 <= (threadIdx_x_1 + 12) % 81 and (threadIdx_x_1 + 12) % 81 < 72 and 1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1470) // 81 * 49 + (threadIdx_x_1 + 12) % 81 // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1568] = T.if_then_else(9 <= (threadIdx_x_1 + 29) % 81 and (threadIdx_x_1 + 29) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1568) // 81 * 49 + (threadIdx_x_1 + 29) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1666] = T.if_then_else(9 <= (threadIdx_x_1 + 46) % 81 and (threadIdx_x_1 + 46) % 81 < 72 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1666) // 81 * 49 + (threadIdx_x_1 + 46) % 81 // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1764] = T.if_then_else(1 <= (threadIdx_x_1 // 9 + 7) % 9 and (threadIdx_x_1 + 63) % 81 < 72 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1764) // 81 * 49 + (threadIdx_x_1 // 9 + 7) % 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1862] = T.if_then_else(9 <= (threadIdx_x_1 + 80) % 81 and (threadIdx_x_1 + 80) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1862) // 81 * 49 + (threadIdx_x_1 + 80) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 1960] = T.if_then_else(9 <= (threadIdx_x_1 + 16) % 81 and (threadIdx_x_1 + 16) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1960) // 81 * 49 + (threadIdx_x_1 + 16) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2058] = T.if_then_else(9 <= (threadIdx_x_1 + 33) % 81 and (threadIdx_x_1 + 33) % 81 < 72 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2058) // 81 * 49 + (threadIdx_x_1 + 33) % 81 // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2156] = T.if_then_else(9 <= (threadIdx_x_1 + 50) % 81 and (threadIdx_x_1 + 50) % 81 < 72 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2156) // 81 * 49 + (threadIdx_x_1 + 50) % 81 // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2254] = T.if_then_else(9 <= (threadIdx_x_1 + 67) % 81 and (threadIdx_x_1 + 67) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2254) // 81 * 49 + (threadIdx_x_1 + 67) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2352] = T.if_then_else(9 <= (threadIdx_x_1 + 3) % 81 and (threadIdx_x_1 + 3) % 81 < 72 and 1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2352) // 81 * 49 + (threadIdx_x_1 + 3) % 81 // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ pad_temp_shared_1[threadIdx_x_1 + 2450] = T.if_then_else(9 <= (threadIdx_x_1 + 20) % 81 and (threadIdx_x_1 + 20) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2450) // 81 * 49 + (threadIdx_x_1 + 20) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 98):
+ if T.likely(threadIdx_x_1 < 44):
+ pad_temp_shared_1[threadIdx_x_1 + 2548] = T.if_then_else(threadIdx_x_1 < 35 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 2548) // 81 * 49 + (threadIdx_x_1 + 37) // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
+ threadIdx_x_2 = T.env_thread("threadIdx.x")
+ kernel_shared_1 = T.Buffer((9216,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 64) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 64) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 128) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 128) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 192] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 36864]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 256) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 256) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 320) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 320) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 384] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 73728]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 448) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 448) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 512) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 512) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 576] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 110592]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 640) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 640) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 704) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 704) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 768] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 147456]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 832) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 832) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 896) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 896) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 960] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 184320]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1024) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1024) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1088) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1088) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1152] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 221184]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1216) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1216) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1280) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1280) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1344] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 258048]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1408) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1408) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1472) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1472) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1536] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 294912]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1600) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1600) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1664) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1664) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1728] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 331776]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1792) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1792) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1856) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1856) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 1920] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 368640]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 1984) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 1984) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2048) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2048) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2112] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 405504]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2176) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2176) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2240) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2240) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2304] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 442368]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2368) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2368) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2432) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2432) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2496] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 479232]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2560) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2560) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2624) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2624) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2688] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 516096]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2752) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2752) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2816) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2816) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[threadIdx_x_1 + 2880] = kernel_1[blockIdx_x // 7 * 589824 + threadIdx_x_1 // 24 * 4608 + cse_var_2 + threadIdx_x_1 % 24 // 3 * 9 + cse_var_1 + threadIdx_x_1 % 3 + 552960]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 2944) // 24 * 24 + (threadIdx_x_1 + 16) % 24 // 3 * 3 + (threadIdx_x_1 + 1) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 2944) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 16) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 1) % 3]
- with T.launch_thread(threadIdx_x_1, 64):
- kernel_shared_1[(threadIdx_x_1 + 3008) // 24 * 24 + (threadIdx_x_1 + 8) % 24 // 3 * 3 + (threadIdx_x_1 + 2) % 3] = kernel_1[blockIdx_x // 7 * 589824 + (threadIdx_x_1 + 3008) // 24 * 4608 + cse_var_2 + (threadIdx_x_1 + 8) % 24 // 3 * 9 + cse_var_1 + (threadIdx_x_1 + 2) % 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 47]
- for i1_inner, i3_inner in T.grid(2, 7):
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x * 147456 + cse_var_1 + threadIdx_x_2]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 98] = kernel_1[blockIdx_x * 147456 + cse_var_1 + threadIdx_x_2 + 98]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 196] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 196) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 196) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 294] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 294) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 6]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 392] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 104]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 490] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 490) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 202) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 588] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 588) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 12]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 686] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 686) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 110]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 784] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 784) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 208) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 882] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 882) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 18]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 980] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 980) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 116]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1078] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1078) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 214) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1176] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1176) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 24]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1274] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1274) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 122]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1372] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1372) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 220) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1470] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1470) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 30]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1568] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1568) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 128]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1666] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1666) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 226) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1764] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1764) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 36]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1862] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1862) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 134]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 1960] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 1960) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 232) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2058] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2058) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 42]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2156] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2156) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 140]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2254] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2254) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 238) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2352] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2352) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 48]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2450] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2450) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 146]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2548] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2548) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 244) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2646] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2646) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 54]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2744] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2744) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 152]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2842] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2842) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 250) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 2940] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 2940) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 60]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3038] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3038) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 158]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3136] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3136) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 256) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3234] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3234) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 66]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3332] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3332) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 164]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3430] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3430) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 262) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3528] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3528) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 72]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3626] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3626) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 170]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3724] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3724) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 268) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3822] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3822) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 78]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 3920] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 3920) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 176]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4018] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4018) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 274) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4116] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4116) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 84]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4214] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4214) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 182]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4312] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4312) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 280) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4410] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4410) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 90]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4508] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4508) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 188]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4606] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4606) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 286) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4704] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4704) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 96]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4802] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4802) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 194) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4900] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4900) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 4]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 4998] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 4998) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 102]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5096] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5096) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 200) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5194] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5194) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 10]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5292] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5292) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 108]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5390] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5390) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 206) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5488] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5488) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 16]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5586] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5586) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 114]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5684] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5684) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 212) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5782] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5782) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 22]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5880] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5880) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 120]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 5978] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 5978) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 218) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6076] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6076) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 28]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6174] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6174) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 126]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6272] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6272) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 224) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6370] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6370) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 34]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6468] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6468) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 132]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6566] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6566) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 230) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6664] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6664) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 40]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6762] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6762) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 138]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6860] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6860) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 236) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 6958] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 6958) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 46]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7056] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7056) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 144]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7154] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7154) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 242) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7252] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7252) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 52]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7350] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7350) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 150]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7448] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7448) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 248) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7546] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7546) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 58]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7644] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7644) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 156]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7742] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7742) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 254) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7840] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7840) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 64]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 7938] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 7938) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 162]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8036] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8036) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 260) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8134] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8134) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 70]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8232] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8232) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 168]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8330] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8330) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 266) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8428] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8428) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 76]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8526] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8526) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 174]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8624] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8624) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 272) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8722] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8722) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 82]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8820] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8820) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 180]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 8918] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 8918) // 288 * 4608 + cse_var_1 + (threadIdx_x_2 + 278) % 288]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 9016] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 9016) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 88]
+ with T.launch_thread(threadIdx_x_2, 98):
+ kernel_shared_1[threadIdx_x_2 + 9114] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 9114) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 186]
+ with T.launch_thread(threadIdx_x_2, 98):
+ if T.likely(threadIdx_x_2 < 4):
+ kernel_shared_1[threadIdx_x_2 + 9212] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 9212) // 288 * 4608 + cse_var_1 + threadIdx_x_2 + 284]
+ for rc_outer_inner in range(16):
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4608]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 9]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4617]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 288]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4896]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 297]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4905]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 576]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5184]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 585]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5193]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 864]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5472]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 873]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5481]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1152]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5760]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1161]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5769]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1440]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6048]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1449]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6057]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1728]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6336]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1737]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6345]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2016]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6624]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2025]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6633]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4609]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 10]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4618]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 289]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4897]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 298]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4906]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 577]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5185]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 586]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5194]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 865]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5473]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 874]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5482]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1153]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5761]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1162]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5770]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1441]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6049]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1450]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6058]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1729]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6337]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1738]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6346]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2017]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6625]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2026]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6634]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4610]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 11]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4619]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 290]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4898]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 299]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4907]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 578]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5186]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 587]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5195]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 866]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5474]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 875]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5483]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1154]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5762]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1163]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5771]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1442]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6050]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1451]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6059]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1730]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6338]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1739]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6347]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2018]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6626]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2027]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6635]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 3]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4611]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 12]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4620]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 291]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4899]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 300]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4908]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 579]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5187]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 588]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5196]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 867]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5475]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 876]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5484]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1155]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5763]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1164]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5772]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1443]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6051]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1452]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6060]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1731]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6339]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1740]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6348]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2019]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 9] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6627]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2028]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 90] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6636]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4612]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 13]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4621]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 292]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4900]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 301]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4909]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 580]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5188]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 589]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5197]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 868]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5476]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 877]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5485]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1156]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5764]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1165]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5773]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1444]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6052]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1453]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6061]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1732]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6340]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1741]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6349]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2020]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 10] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6628]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2029]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 91] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6637]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4613]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 14]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4622]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 293]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4901]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 302]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4910]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 581]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5189]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 590]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5198]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 869]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5477]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 878]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5486]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1157]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5765]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1166]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5774]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1445]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6053]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1454]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6062]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1733]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6341]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1742]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6350]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2021]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 11] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6629]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2030]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 92] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6638]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4614]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 15]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4623]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 294]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4902]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 303]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4911]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 582]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5190]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 591]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5199]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 870]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5478]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 879]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5487]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1158]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5766]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1167]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5775]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1446]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6054]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1455]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6063]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1734]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6342]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1743]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6351]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2022]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 18] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6630]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2031]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 99] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6639]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 7]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4615]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 16]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4624]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 295]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4903]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 304]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4912]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 583]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5191]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 592]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5200]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 871]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5479]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 880]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5488]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1159]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5767]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1168]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5776]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1447]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6055]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1456]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6064]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1735]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6343]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1744]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6352]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2023]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 19] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6631]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2032]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 100] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6640]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 8]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4616]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 17]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4625]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 296]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4904]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 305]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 4913]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 584]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5192]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 593]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5201]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 872]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5480]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 881]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5489]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1160]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5768]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1169]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 5777]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1448]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6056]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1457]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6065]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1736]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6344]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 1745]
+ conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6353]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2024]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 20] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6632]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 2033]
+ conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + threadIdx_x % 7 + 101] * kernel_shared_1[threadIdx_x // 49 * 2304 + rc_outer_inner * 18 + 6641]
+ for i1_inner in range(8):
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x // 7 * 6272 + threadIdx_x * 98 + i1_inner * 49 + blockIdx_x % 7 * 7 + i3_inner] = T.max(conv2d_nchw_1[i1_inner * 7 + i3_inner] + bias_1[blockIdx_x // 7 * 128 + threadIdx_x * 2 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 1568 + threadIdx_x // 49 * 392 + i1_inner * 49 + threadIdx_x % 49] = T.max(conv2d_nchw_1[i1_inner] + bias_1[blockIdx_x * 32 + threadIdx_x // 49 * 8 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 1568 + threadIdx_x // 49 * 392 + i1_inner * 49 + threadIdx_x % 49 + 784] = T.max(conv2d_nchw_1[i1_inner + 8] + bias_1[blockIdx_x * 32 + threadIdx_x // 49 * 8 + i1_inner + 16], T.float32(0))
</pre></div>
</div>
</div>
@@ -1019,7 +1114,7 @@ class Module:
<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: 0.339 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.268 ms
</pre></div>
</div>
</div>
@@ -1049,35 +1144,35 @@ 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=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=64)
-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=8)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=2)
+conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=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=1)
+conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
-conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
-conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, 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=2)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=16)
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=1)
+conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
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=2)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
-compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=8)
+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=2)
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=1)
+compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
-compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
-compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_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)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -1097,14 +1192,14 @@ 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=64)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=4)
+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=64)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
-s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+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)
CUDA source code:
@@ -1129,431 +1224,450 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[14];
- __shared__ float pad_temp_shared[72];
- __shared__ float kernel_shared[3072];
+extern "C" __global__ void __launch_bounds__(98) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[16];
+ __shared__ float pad_temp_shared[2592];
+ __shared__ float kernel_shared[9216];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
conv2d_nchw[7] = 0.000000e+00f;
- conv2d_nchw[8] = 0.000000e+00f;
- conv2d_nchw[9] = 0.000000e+00f;
- conv2d_nchw[10] = 0.000000e+00f;
- conv2d_nchw[11] = 0.000000e+00f;
- conv2d_nchw[12] = 0.000000e+00f;
- conv2d_nchw[13] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
- for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
- __syncthreads();
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[(((int)threadIdx.x) * 4)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 4) % 9))) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) * 4) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 1)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 1) % 9))) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 1) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 2)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 2) % 9))) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 2) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 3)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 3) % 9))) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 3) % 9)) - 8)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 64) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 128) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((((((int)threadIdx.x) + 256) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 320) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((((((int)threadIdx.x) + 448) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 512) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((((((int)threadIdx.x) + 640) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 704) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
- kernel_shared[(((((((int)threadIdx.x) + 832) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 896) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 896) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
- kernel_shared[(((((((int)threadIdx.x) + 1024) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1088) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
- kernel_shared[(((((((int)threadIdx.x) + 1216) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1280) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
- kernel_shared[(((((((int)threadIdx.x) + 1408) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1472) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 294912)];
- kernel_shared[(((((((int)threadIdx.x) + 1600) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1664) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 331776)];
- kernel_shared[(((((((int)threadIdx.x) + 1792) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1856) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 368640)];
- kernel_shared[(((((((int)threadIdx.x) + 1984) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2048) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 405504)];
- kernel_shared[(((((((int)threadIdx.x) + 2176) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2240) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 442368)];
- kernel_shared[(((((((int)threadIdx.x) + 2368) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2432) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 479232)];
- kernel_shared[(((((((int)threadIdx.x) + 2560) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2624) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 516096)];
- kernel_shared[(((((((int)threadIdx.x) + 2752) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2816) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 552960)];
- kernel_shared[(((((((int)threadIdx.x) + 2944) / 24) * 24) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 3008) / 24) * 24) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+ conv2d_nchw[15] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[((int)threadIdx.x)] = (((((9 <= (((int)threadIdx.x) % 81)) && ((((int)threadIdx.x) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 81) * 49)) + (((((int)threadIdx.x) % 81) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 98)] = (((((9 <= ((((int)threadIdx.x) + 17) % 81)) && (((((int)threadIdx.x) + 17) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 98) / 81) * 49)) + ((((((int)threadIdx.x) + 17) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((9 <= ((((int)threadIdx.x) + 34) % 81)) && (((((int)threadIdx.x) + 34) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 196) / 81) * 49)) + ((((((int)threadIdx.x) + 34) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 294)] = (((((9 <= ((((int)threadIdx.x) + 51) % 81)) && (((((int)threadIdx.x) + 51) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 294) / 81) * 49)) + ((((((int)threadIdx.x) + 51) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((9 <= ((((int)threadIdx.x) + 68) % 81)) && (((((int)threadIdx.x) + 68) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 392) / 81) * 49)) + ((((((int)threadIdx.x) + 68) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 490)] = (((((9 <= ((((int)threadIdx.x) + 4) % 81)) && (((((int)threadIdx.x) + 4) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 490) / 81) * 49)) + ((((((int)threadIdx.x) + 4) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 588)] = (((((9 <= ((((int)threadIdx.x) + 21) % 81)) && (((((int)threadIdx.x) + 21) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 588) / 81) * 49)) + ((((((int)threadIdx.x) + 21) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 686)] = (((((9 <= ((((int)threadIdx.x) + 38) % 81)) && (((((int)threadIdx.x) + 38) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 686) / 81) * 49)) + ((((((int)threadIdx.x) + 38) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((9 <= ((((int)threadIdx.x) + 55) % 81)) && (((((int)threadIdx.x) + 55) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 784) / 81) * 49)) + ((((((int)threadIdx.x) + 55) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 882)] = (((((1 <= (((((int)threadIdx.x) / 9) + 8) % 9)) && (((((int)threadIdx.x) + 72) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 882) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 8) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 980)] = (((((9 <= ((((int)threadIdx.x) + 8) % 81)) && (((((int)threadIdx.x) + 8) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 980) / 81) * 49)) + ((((((int)threadIdx.x) + 8) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1078)] = (((((9 <= ((((int)threadIdx.x) + 25) % 81)) && (((((int)threadIdx.x) + 25) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1078) / 81) * 49)) + ((((((int)threadIdx.x) + 25) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((9 <= ((((int)threadIdx.x) + 42) % 81)) && (((((int)threadIdx.x) + 42) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1176) / 81) * 49)) + ((((((int)threadIdx.x) + 42) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1274)] = (((((9 <= ((((int)threadIdx.x) + 59) % 81)) && (((((int)threadIdx.x) + 59) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1274) / 81) * 49)) + ((((((int)threadIdx.x) + 59) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1372)] = (((((9 <= ((((int)threadIdx.x) + 76) % 81)) && (((((int)threadIdx.x) + 76) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1372) / 81) * 49)) + ((((((int)threadIdx.x) + 76) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1470)] = (((((9 <= ((((int)threadIdx.x) + 12) % 81)) && (((((int)threadIdx.x) + 12) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1470) / 81) * 49)) + ((((((int)threadIdx.x) + 12) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1568)] = (((((9 <= ((((int)threadIdx.x) + 29) % 81)) && (((((int)threadIdx.x) + 29) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1568) / 81) * 49)) + ((((((int)threadIdx.x) + 29) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1666)] = (((((9 <= ((((int)threadIdx.x) + 46) % 81)) && (((((int)threadIdx.x) + 46) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1666) / 81) * 49)) + ((((((int)threadIdx.x) + 46) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1764)] = (((((1 <= (((((int)threadIdx.x) / 9) + 7) % 9)) && (((((int)threadIdx.x) + 63) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1764) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 7) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1862)] = (((((9 <= ((((int)threadIdx.x) + 80) % 81)) && (((((int)threadIdx.x) + 80) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1862) / 81) * 49)) + ((((((int)threadIdx.x) + 80) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1960)] = (((((9 <= ((((int)threadIdx.x) + 16) % 81)) && (((((int)threadIdx.x) + 16) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1960) / 81) * 49)) + ((((((int)threadIdx.x) + 16) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2058)] = (((((9 <= ((((int)threadIdx.x) + 33) % 81)) && (((((int)threadIdx.x) + 33) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2058) / 81) * 49)) + ((((((int)threadIdx.x) + 33) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2156)] = (((((9 <= ((((int)threadIdx.x) + 50) % 81)) && (((((int)threadIdx.x) + 50) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2156) / 81) * 49)) + ((((((int)threadIdx.x) + 50) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2254)] = (((((9 <= ((((int)threadIdx.x) + 67) % 81)) && (((((int)threadIdx.x) + 67) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2254) / 81) * 49)) + ((((((int)threadIdx.x) + 67) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2352)] = (((((9 <= ((((int)threadIdx.x) + 3) % 81)) && (((((int)threadIdx.x) + 3) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2352) / 81) * 49)) + ((((((int)threadIdx.x) + 3) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2450)] = (((((9 <= ((((int)threadIdx.x) + 20) % 81)) && (((((int)threadIdx.x) + 20) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2450) / 81) * 49)) + ((((((int)threadIdx.x) + 20) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 44) {
+ pad_temp_shared[(((int)threadIdx.x) + 2548)] = ((((((int)threadIdx.x) < 35) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 2548) / 81) * 49)) + (((((int)threadIdx.x) + 37) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
}
- }
- for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 147456) + (rc_outer_outer * 288)) + ((int)threadIdx.x))];
+ kernel_shared[(((int)threadIdx.x) + 98)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 98)];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 196) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 196) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 294)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 294) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 6)];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 104)];
+ kernel_shared[(((int)threadIdx.x) + 490)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 490) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 202) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 588)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 588) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 12)];
+ kernel_shared[(((int)threadIdx.x) + 686)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 686) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 110)];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 208) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 882)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 882) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 18)];
+ kernel_shared[(((int)threadIdx.x) + 980)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 980) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 116)];
+ kernel_shared[(((int)threadIdx.x) + 1078)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1078) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 214) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1176) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 1274)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1274) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 122)];
+ kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1372) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 220) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1470)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1470) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 30)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1568) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 128)];
+ kernel_shared[(((int)threadIdx.x) + 1666)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1666) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 226) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 1764)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1764) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 36)];
+ kernel_shared[(((int)threadIdx.x) + 1862)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1862) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 134)];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1960) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 232) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2058)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2058) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 42)];
+ kernel_shared[(((int)threadIdx.x) + 2156)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2156) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 140)];
+ kernel_shared[(((int)threadIdx.x) + 2254)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2254) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 238) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2352) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 2450)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2450) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 146)];
+ kernel_shared[(((int)threadIdx.x) + 2548)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2548) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 244) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2646)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2646) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 54)];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2744) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 152)];
+ kernel_shared[(((int)threadIdx.x) + 2842)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2842) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 250) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 2940)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2940) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 60)];
+ kernel_shared[(((int)threadIdx.x) + 3038)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3038) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 158)];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3136) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 256) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3234)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3234) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 66)];
+ kernel_shared[(((int)threadIdx.x) + 3332)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3332) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 164)];
+ kernel_shared[(((int)threadIdx.x) + 3430)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3430) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 262) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3528)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3528) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 3626)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3626) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 170)];
+ kernel_shared[(((int)threadIdx.x) + 3724)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3724) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 268) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 3822)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3822) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 78)];
+ kernel_shared[(((int)threadIdx.x) + 3920)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3920) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 176)];
+ kernel_shared[(((int)threadIdx.x) + 4018)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4018) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 274) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4116)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4116) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 84)];
+ kernel_shared[(((int)threadIdx.x) + 4214)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4214) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 182)];
+ kernel_shared[(((int)threadIdx.x) + 4312)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4312) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 280) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4410)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4410) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 90)];
+ kernel_shared[(((int)threadIdx.x) + 4508)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4508) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 188)];
+ kernel_shared[(((int)threadIdx.x) + 4606)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4606) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 286) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4704)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4704) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 96)];
+ kernel_shared[(((int)threadIdx.x) + 4802)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4802) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 194) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 4900)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4900) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 4)];
+ kernel_shared[(((int)threadIdx.x) + 4998)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4998) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 102)];
+ kernel_shared[(((int)threadIdx.x) + 5096)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5096) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 200) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 5194)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5194) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 10)];
+ kernel_shared[(((int)threadIdx.x) + 5292)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5292) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 108)];
+ kernel_shared[(((int)threadIdx.x) + 5390)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5390) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 206) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 5488)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5488) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 16)];
+ kernel_shared[(((int)threadIdx.x) + 5586)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5586) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 114)];
+ kernel_shared[(((int)threadIdx.x) + 5684)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5684) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 212) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 5782)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5782) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 22)];
+ kernel_shared[(((int)threadIdx.x) + 5880)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5880) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 120)];
+ kernel_shared[(((int)threadIdx.x) + 5978)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5978) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 218) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 6076)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6076) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 28)];
+ kernel_shared[(((int)threadIdx.x) + 6174)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6174) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 126)];
+ kernel_shared[(((int)threadIdx.x) + 6272)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6272) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 224) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 6370)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6370) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 34)];
+ kernel_shared[(((int)threadIdx.x) + 6468)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6468) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 132)];
+ kernel_shared[(((int)threadIdx.x) + 6566)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6566) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 230) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 6664)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6664) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 40)];
+ kernel_shared[(((int)threadIdx.x) + 6762)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6762) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 138)];
+ kernel_shared[(((int)threadIdx.x) + 6860)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6860) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 236) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 6958)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6958) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 46)];
+ kernel_shared[(((int)threadIdx.x) + 7056)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7056) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 7154)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7154) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 242) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 7252)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7252) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 52)];
+ kernel_shared[(((int)threadIdx.x) + 7350)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7350) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 150)];
+ kernel_shared[(((int)threadIdx.x) + 7448)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7448) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 248) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 7546)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7546) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 58)];
+ kernel_shared[(((int)threadIdx.x) + 7644)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7644) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 156)];
+ kernel_shared[(((int)threadIdx.x) + 7742)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7742) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 254) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 7840)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7840) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 64)];
+ kernel_shared[(((int)threadIdx.x) + 7938)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 7938) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 162)];
+ kernel_shared[(((int)threadIdx.x) + 8036)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8036) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 260) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 8134)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8134) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 70)];
+ kernel_shared[(((int)threadIdx.x) + 8232)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8232) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 168)];
+ kernel_shared[(((int)threadIdx.x) + 8330)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8330) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 266) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 8428)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8428) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 76)];
+ kernel_shared[(((int)threadIdx.x) + 8526)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8526) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 174)];
+ kernel_shared[(((int)threadIdx.x) + 8624)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8624) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 272) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 8722)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8722) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 82)];
+ kernel_shared[(((int)threadIdx.x) + 8820)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8820) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 180)];
+ kernel_shared[(((int)threadIdx.x) + 8918)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 8918) / 288) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) + 278) % 288))];
+ kernel_shared[(((int)threadIdx.x) + 9016)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 9016) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 88)];
+ kernel_shared[(((int)threadIdx.x) + 9114)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 9114) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 186)];
+ if (((int)threadIdx.x) < 4) {
+ kernel_shared[(((int)threadIdx.x) + 9212)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 9212) / 288) * 4608)) + (rc_outer_outer * 288)) + ((int)threadIdx.x)) + 284)];
+ }
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 16; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18))]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4608)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 9)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4617)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 288)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4896)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 297)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4905)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 576)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5184)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 585)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5193)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 864)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5472)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 873)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5481)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1152)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5760)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1161)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5769)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1440)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6048)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1449)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6057)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1728)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6336)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1737)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6345)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2016)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6624)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2025)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6633)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4609)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 10)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4618)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 289)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4897)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 298)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4906)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 577)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5185)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 586)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5194)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 865)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5473)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 874)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5482)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1153)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5761)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1162)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5770)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1441)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6049)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1450)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6058)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1729)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6337)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1738)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6346)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2017)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6625)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2026)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6634)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4610)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 11)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4619)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 290)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4898)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 299)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4907)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 578)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5186)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 587)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5195)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 866)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5474)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 875)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5483)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1154)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5762)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1163)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5771)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1442)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6050)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1451)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6059)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1730)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6338)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1739)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6347)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2018)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6626)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2027)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6635)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 3)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4611)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 12)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4620)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 291)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4899)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 300)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4908)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 579)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5187)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 588)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5196)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 867)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5475)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 876)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5484)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1155)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5763)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1164)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5772)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1443)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6051)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1452)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6060)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1731)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6339)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1740)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6348)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2019)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6627)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2028)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6636)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4612)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 13)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4621)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 292)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4900)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 301)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4909)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 580)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5188)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 589)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5197)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 868)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5476)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 877)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5485)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1156)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5764)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1165)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5773)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1444)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6052)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1453)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6061)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1732)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6340)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1741)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6349)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2020)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6628)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2029)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6637)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4613)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 14)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4622)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 293)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4901)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 302)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4910)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 581)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5189)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 590)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5198)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 869)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5477)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 878)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5486)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1157)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5765)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1166)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5774)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1445)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6053)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1454)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6062)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1733)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6341)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1742)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6350)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2021)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6629)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2030)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6638)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4614)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 15)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4623)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 294)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4902)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 303)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4911)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 582)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5190)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 591)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5199)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 870)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5478)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 879)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5487)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1158)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5766)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1167)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5775)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1446)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6054)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1455)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6063)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1734)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6342)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1743)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6351)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2022)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6630)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2031)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6639)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 7)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4615)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 16)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4624)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 295)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4903)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 304)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4912)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 583)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5191)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 592)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5200)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 871)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5479)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 880)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5488)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1159)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5767)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1168)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5776)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1447)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6055)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1456)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6064)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1735)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6343)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1744)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6352)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2023)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6631)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2032)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6640)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 8)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4616)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 17)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4625)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 296)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4904)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 305)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 4913)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 584)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5192)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 593)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5201)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 872)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5480)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 881)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5489)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1160)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5768)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1169)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 5777)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1448)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6056)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1457)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6065)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1736)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6344)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 1745)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6353)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2024)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6632)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 2033)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 49) * 2304) + (rc_outer_inner * 18)) + 6641)]));
}
}
+ for (int i1_inner = 0; i1_inner < 8; ++i1_inner) {
+ compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 392)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 8)) + i1_inner)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 392)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 784)] = max((conv2d_nchw[(i1_inner + 8)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 8)) + i1_inner) + 16)]), 0.000000e+00f);
+ }
}
</pre></div>
</div>
@@ -1585,9 +1699,10 @@ In the example below we resume the status and do more 5 trials.</p>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Resume search:
Get devices for measurement successfully!
+.T
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes 57.183 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes 51.242 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 d4657256c1..b953166641 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -921,7 +921,7 @@ 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)
- 8.1218 8.1208 8.1314 8.1133 0.0074
+ 8.1176 8.1192 8.1229 8.1106 0.0052
</pre></div>
</div>
</div>
@@ -943,7 +943,7 @@ 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 18.112 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 14.798 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-network-cuda-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/eafe360d52540634c9eea0fa89e804bd/tune_network_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_network_cuda.py</span></code></a></p>
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 173588509d..1e8a29c7ba 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -940,7 +940,7 @@ 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)
- 798.0223 799.0309 802.9418 792.0942 4.4856
+ 758.5152 758.9505 759.4899 757.1053 1.0210
</pre></div>
</div>
</div>
@@ -962,7 +962,7 @@ 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 53.239 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 47.856 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 bd76be9758..7e20b601be 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -637,26 +637,407 @@ class Module:
@T.prim_func
def main(placeholder: T.Buffer((128, 256), "float32"), placeholder_1: T.Buffer((4916, 16, 1), "float32"), placeholder_2: T.Buffer((4916,), "int32"), placeholder_3: T.Buffer((33,), "int32"), placeholder_4: T.Buffer((128, 512), "float32"), compute: T.Buffer((128, 512), "float32")):
T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
- for i0_outer_i1_outer_fused in T.parallel(64):
- compute_1 = T.allocate([1024], "float32", "global")
- compute_2 = T.Buffer((1024,), data=compute_1)
- for nb_j_inner in range(2):
- for i_inner_init, j_init in T.grid(32, 16):
- compute_2[i_inner_init * 32 + nb_j_inner * 16 + j_init] = T.float32(0)
- for elem_idx, i_inner, j in T.grid(T.Let(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1], where={cse_var_1: i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner}), 32, 16):
- cse_var_1 = T.int32()
+ for i0_outer_i1_outer_fused in T.parallel(128):
+ compute_1 = T.allocate([512], "float32", "global")
+ compute_2 = T.Buffer((512,), data=compute_1)
+ for i_outer_inner, nb_j_inner in T.grid(2, 2):
+ cse_var_2: T.int32 = i_outer_inner * 256 + nb_j_inner * 16
+ cse_var_1: T.int32 = i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner
+ compute_2[cse_var_2] = T.float32(0)
+ compute_2[cse_var_2 + 1] = T.float32(0)
+ compute_2[cse_var_2 + 2] = T.float32(0)
+ compute_2[cse_var_2 + 3] = T.float32(0)
+ compute_2[cse_var_2 + 4] = T.float32(0)
+ compute_2[cse_var_2 + 5] = T.float32(0)
+ compute_2[cse_var_2 + 6] = T.float32(0)
+ compute_2[cse_var_2 + 7] = T.float32(0)
+ compute_2[cse_var_2 + 8] = T.float32(0)
+ compute_2[cse_var_2 + 9] = T.float32(0)
+ compute_2[cse_var_2 + 10] = T.float32(0)
+ compute_2[cse_var_2 + 11] = T.float32(0)
+ compute_2[cse_var_2 + 12] = T.float32(0)
+ compute_2[cse_var_2 + 13] = T.float32(0)
+ compute_2[cse_var_2 + 14] = T.float32(0)
+ compute_2[cse_var_2 + 15] = T.float32(0)
+ compute_2[cse_var_2 + 32] = T.float32(0)
+ compute_2[cse_var_2 + 33] = T.float32(0)
+ compute_2[cse_var_2 + 34] = T.float32(0)
+ compute_2[cse_var_2 + 35] = T.float32(0)
+ compute_2[cse_var_2 + 36] = T.float32(0)
+ compute_2[cse_var_2 + 37] = T.float32(0)
+ compute_2[cse_var_2 + 38] = T.float32(0)
+ compute_2[cse_var_2 + 39] = T.float32(0)
+ compute_2[cse_var_2 + 40] = T.float32(0)
+ compute_2[cse_var_2 + 41] = T.float32(0)
+ compute_2[cse_var_2 + 42] = T.float32(0)
+ compute_2[cse_var_2 + 43] = T.float32(0)
+ compute_2[cse_var_2 + 44] = T.float32(0)
+ compute_2[cse_var_2 + 45] = T.float32(0)
+ compute_2[cse_var_2 + 46] = T.float32(0)
+ compute_2[cse_var_2 + 47] = T.float32(0)
+ compute_2[cse_var_2 + 64] = T.float32(0)
+ compute_2[cse_var_2 + 65] = T.float32(0)
+ compute_2[cse_var_2 + 66] = T.float32(0)
+ compute_2[cse_var_2 + 67] = T.float32(0)
+ compute_2[cse_var_2 + 68] = T.float32(0)
+ compute_2[cse_var_2 + 69] = T.float32(0)
+ compute_2[cse_var_2 + 70] = T.float32(0)
+ compute_2[cse_var_2 + 71] = T.float32(0)
+ compute_2[cse_var_2 + 72] = T.float32(0)
+ compute_2[cse_var_2 + 73] = T.float32(0)
+ compute_2[cse_var_2 + 74] = T.float32(0)
+ compute_2[cse_var_2 + 75] = T.float32(0)
+ compute_2[cse_var_2 + 76] = T.float32(0)
+ compute_2[cse_var_2 + 77] = T.float32(0)
+ compute_2[cse_var_2 + 78] = T.float32(0)
+ compute_2[cse_var_2 + 79] = T.float32(0)
+ compute_2[cse_var_2 + 96] = T.float32(0)
+ compute_2[cse_var_2 + 97] = T.float32(0)
+ compute_2[cse_var_2 + 98] = T.float32(0)
+ compute_2[cse_var_2 + 99] = T.float32(0)
+ compute_2[cse_var_2 + 100] = T.float32(0)
+ compute_2[cse_var_2 + 101] = T.float32(0)
+ compute_2[cse_var_2 + 102] = T.float32(0)
+ compute_2[cse_var_2 + 103] = T.float32(0)
+ compute_2[cse_var_2 + 104] = T.float32(0)
+ compute_2[cse_var_2 + 105] = T.float32(0)
+ compute_2[cse_var_2 + 106] = T.float32(0)
+ compute_2[cse_var_2 + 107] = T.float32(0)
+ compute_2[cse_var_2 + 108] = T.float32(0)
+ compute_2[cse_var_2 + 109] = T.float32(0)
+ compute_2[cse_var_2 + 110] = T.float32(0)
+ compute_2[cse_var_2 + 111] = T.float32(0)
+ compute_2[cse_var_2 + 128] = T.float32(0)
+ compute_2[cse_var_2 + 129] = T.float32(0)
+ compute_2[cse_var_2 + 130] = T.float32(0)
+ compute_2[cse_var_2 + 131] = T.float32(0)
+ compute_2[cse_var_2 + 132] = T.float32(0)
+ compute_2[cse_var_2 + 133] = T.float32(0)
+ compute_2[cse_var_2 + 134] = T.float32(0)
+ compute_2[cse_var_2 + 135] = T.float32(0)
+ compute_2[cse_var_2 + 136] = T.float32(0)
+ compute_2[cse_var_2 + 137] = T.float32(0)
+ compute_2[cse_var_2 + 138] = T.float32(0)
+ compute_2[cse_var_2 + 139] = T.float32(0)
+ compute_2[cse_var_2 + 140] = T.float32(0)
+ compute_2[cse_var_2 + 141] = T.float32(0)
+ compute_2[cse_var_2 + 142] = T.float32(0)
+ compute_2[cse_var_2 + 143] = T.float32(0)
+ compute_2[cse_var_2 + 160] = T.float32(0)
+ compute_2[cse_var_2 + 161] = T.float32(0)
+ compute_2[cse_var_2 + 162] = T.float32(0)
+ compute_2[cse_var_2 + 163] = T.float32(0)
+ compute_2[cse_var_2 + 164] = T.float32(0)
+ compute_2[cse_var_2 + 165] = T.float32(0)
+ compute_2[cse_var_2 + 166] = T.float32(0)
+ compute_2[cse_var_2 + 167] = T.float32(0)
+ compute_2[cse_var_2 + 168] = T.float32(0)
+ compute_2[cse_var_2 + 169] = T.float32(0)
+ compute_2[cse_var_2 + 170] = T.float32(0)
+ compute_2[cse_var_2 + 171] = T.float32(0)
+ compute_2[cse_var_2 + 172] = T.float32(0)
+ compute_2[cse_var_2 + 173] = T.float32(0)
+ compute_2[cse_var_2 + 174] = T.float32(0)
+ compute_2[cse_var_2 + 175] = T.float32(0)
+ compute_2[cse_var_2 + 192] = T.float32(0)
+ compute_2[cse_var_2 + 193] = T.float32(0)
+ compute_2[cse_var_2 + 194] = T.float32(0)
+ compute_2[cse_var_2 + 195] = T.float32(0)
+ compute_2[cse_var_2 + 196] = T.float32(0)
+ compute_2[cse_var_2 + 197] = T.float32(0)
+ compute_2[cse_var_2 + 198] = T.float32(0)
+ compute_2[cse_var_2 + 199] = T.float32(0)
+ compute_2[cse_var_2 + 200] = T.float32(0)
+ compute_2[cse_var_2 + 201] = T.float32(0)
+ compute_2[cse_var_2 + 202] = T.float32(0)
+ compute_2[cse_var_2 + 203] = T.float32(0)
+ compute_2[cse_var_2 + 204] = T.float32(0)
+ compute_2[cse_var_2 + 205] = T.float32(0)
+ compute_2[cse_var_2 + 206] = T.float32(0)
+ compute_2[cse_var_2 + 207] = T.float32(0)
+ compute_2[cse_var_2 + 224] = T.float32(0)
+ compute_2[cse_var_2 + 225] = T.float32(0)
+ compute_2[cse_var_2 + 226] = T.float32(0)
+ compute_2[cse_var_2 + 227] = T.float32(0)
+ compute_2[cse_var_2 + 228] = T.float32(0)
+ compute_2[cse_var_2 + 229] = T.float32(0)
+ compute_2[cse_var_2 + 230] = T.float32(0)
+ compute_2[cse_var_2 + 231] = T.float32(0)
+ compute_2[cse_var_2 + 232] = T.float32(0)
+ compute_2[cse_var_2 + 233] = T.float32(0)
+ compute_2[cse_var_2 + 234] = T.float32(0)
+ compute_2[cse_var_2 + 235] = T.float32(0)
+ compute_2[cse_var_2 + 236] = T.float32(0)
+ compute_2[cse_var_2 + 237] = T.float32(0)
+ compute_2[cse_var_2 + 238] = T.float32(0)
+ compute_2[cse_var_2 + 239] = T.float32(0)
+ for elem_idx in range(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_3: T.int32 = i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner
- cse_var_2: T.int32 = i_inner * 32 + nb_j_inner * 16 + j
+ cse_var_131: T.int32 = elem_idx * 16
+ cse_var_130: T.int32 = cse_var_2 + 99
+ cse_var_129: T.int32 = cse_var_2 + 98
+ cse_var_128: T.int32 = cse_var_2 + 97
+ cse_var_127: T.int32 = cse_var_2 + 96
+ cse_var_126: T.int32 = cse_var_2 + 9
+ cse_var_125: T.int32 = cse_var_2 + 8
+ cse_var_124: T.int32 = cse_var_2 + 79
+ cse_var_123: T.int32 = cse_var_2 + 78
+ cse_var_122: T.int32 = cse_var_2 + 77
+ cse_var_121: T.int32 = cse_var_2 + 76
+ cse_var_120: T.int32 = cse_var_2 + 75
+ cse_var_119: T.int32 = cse_var_2 + 74
+ cse_var_118: T.int32 = cse_var_2 + 73
+ cse_var_117: T.int32 = cse_var_2 + 72
+ cse_var_116: T.int32 = cse_var_2 + 71
+ cse_var_115: T.int32 = cse_var_2 + 70
+ cse_var_114: T.int32 = cse_var_2 + 7
+ cse_var_113: T.int32 = cse_var_2 + 69
+ cse_var_112: T.int32 = cse_var_2 + 68
+ cse_var_111: T.int32 = cse_var_2 + 67
+ cse_var_110: T.int32 = cse_var_2 + 66
+ cse_var_109: T.int32 = cse_var_2 + 65
+ cse_var_108: T.int32 = cse_var_2 + 64
+ cse_var_107: T.int32 = cse_var_2 + 6
+ cse_var_106: T.int32 = cse_var_2 + 5
+ cse_var_105: T.int32 = cse_var_2 + 47
+ cse_var_104: T.int32 = cse_var_2 + 46
+ cse_var_103: T.int32 = cse_var_2 + 45
+ cse_var_102: T.int32 = cse_var_2 + 44
+ cse_var_101: T.int32 = cse_var_2 + 43
+ cse_var_100: T.int32 = cse_var_2 + 42
+ cse_var_99: T.int32 = cse_var_2 + 41
+ cse_var_98: T.int32 = cse_var_2 + 40
+ cse_var_97: T.int32 = cse_var_2 + 4
+ cse_var_96: T.int32 = cse_var_2 + 39
+ cse_var_95: T.int32 = cse_var_2 + 38
+ cse_var_94: T.int32 = cse_var_2 + 37
+ cse_var_93: T.int32 = cse_var_2 + 36
+ cse_var_92: T.int32 = cse_var_2 + 35
+ cse_var_91: T.int32 = cse_var_2 + 34
+ cse_var_90: T.int32 = cse_var_2 + 33
+ cse_var_89: T.int32 = cse_var_2 + 32
+ cse_var_88: T.int32 = cse_var_2 + 3
+ cse_var_87: T.int32 = cse_var_2 + 239
+ cse_var_86: T.int32 = cse_var_2 + 238
+ cse_var_85: T.int32 = cse_var_2 + 237
+ cse_var_84: T.int32 = cse_var_2 + 236
+ cse_var_83: T.int32 = cse_var_2 + 235
+ cse_var_82: T.int32 = cse_var_2 + 234
+ cse_var_81: T.int32 = cse_var_2 + 233
+ cse_var_80: T.int32 = cse_var_2 + 232
+ cse_var_79: T.int32 = cse_var_2 + 231
+ cse_var_78: T.int32 = cse_var_2 + 230
+ cse_var_77: T.int32 = cse_var_2 + 229
+ cse_var_76: T.int32 = cse_var_2 + 228
+ cse_var_75: T.int32 = cse_var_2 + 227
+ cse_var_74: T.int32 = cse_var_2 + 226
+ cse_var_73: T.int32 = cse_var_2 + 225
+ cse_var_72: T.int32 = cse_var_2 + 224
+ cse_var_71: T.int32 = cse_var_2 + 207
+ cse_var_70: T.int32 = cse_var_2 + 206
+ cse_var_69: T.int32 = cse_var_2 + 205
+ cse_var_68: T.int32 = cse_var_2 + 204
+ cse_var_67: T.int32 = cse_var_2 + 203
+ cse_var_66: T.int32 = cse_var_2 + 202
+ cse_var_65: T.int32 = cse_var_2 + 201
+ cse_var_64: T.int32 = cse_var_2 + 200
+ cse_var_63: T.int32 = cse_var_2 + 2
+ cse_var_62: T.int32 = cse_var_2 + 199
+ cse_var_61: T.int32 = cse_var_2 + 198
+ cse_var_60: T.int32 = cse_var_2 + 197
+ cse_var_59: T.int32 = cse_var_2 + 196
+ cse_var_58: T.int32 = cse_var_2 + 195
+ cse_var_57: T.int32 = cse_var_2 + 194
+ cse_var_56: T.int32 = cse_var_2 + 193
+ cse_var_55: T.int32 = cse_var_2 + 192
+ cse_var_54: T.int32 = cse_var_2 + 175
+ cse_var_53: T.int32 = cse_var_2 + 174
+ cse_var_52: T.int32 = cse_var_2 + 173
+ cse_var_51: T.int32 = cse_var_2 + 172
+ cse_var_50: T.int32 = cse_var_2 + 171
+ cse_var_49: T.int32 = cse_var_2 + 170
+ cse_var_48: T.int32 = cse_var_2 + 169
+ cse_var_47: T.int32 = cse_var_2 + 168
+ cse_var_46: T.int32 = cse_var_2 + 167
+ cse_var_45: T.int32 = cse_var_2 + 166
+ cse_var_44: T.int32 = cse_var_2 + 165
+ cse_var_43: T.int32 = cse_var_2 + 164
+ cse_var_42: T.int32 = cse_var_2 + 163
+ cse_var_41: T.int32 = cse_var_2 + 162
+ cse_var_40: T.int32 = cse_var_2 + 161
+ cse_var_39: T.int32 = cse_var_2 + 160
+ cse_var_38: T.int32 = cse_var_2 + 15
+ cse_var_37: T.int32 = cse_var_2 + 143
+ cse_var_36: T.int32 = cse_var_2 + 142
+ cse_var_35: T.int32 = cse_var_2 + 141
+ cse_var_34: T.int32 = cse_var_2 + 140
+ cse_var_33: T.int32 = cse_var_2 + 14
+ cse_var_32: T.int32 = cse_var_2 + 139
+ cse_var_31: T.int32 = cse_var_2 + 138
+ cse_var_30: T.int32 = cse_var_2 + 137
+ cse_var_29: T.int32 = cse_var_2 + 136
+ cse_var_28: T.int32 = cse_var_2 + 135
+ cse_var_27: T.int32 = cse_var_2 + 134
+ cse_var_26: T.int32 = cse_var_2 + 133
+ cse_var_25: T.int32 = cse_var_2 + 132
+ cse_var_24: T.int32 = cse_var_2 + 131
+ cse_var_23: T.int32 = cse_var_2 + 130
+ cse_var_22: T.int32 = cse_var_2 + 13
+ cse_var_21: T.int32 = cse_var_2 + 129
+ cse_var_20: T.int32 = cse_var_2 + 128
+ cse_var_19: T.int32 = cse_var_2 + 12
+ cse_var_18: T.int32 = cse_var_2 + 111
+ cse_var_17: T.int32 = cse_var_2 + 110
+ cse_var_16: T.int32 = cse_var_2 + 11
+ cse_var_15: T.int32 = cse_var_2 + 109
+ cse_var_14: T.int32 = cse_var_2 + 108
+ cse_var_13: T.int32 = cse_var_2 + 107
+ cse_var_12: T.int32 = cse_var_2 + 106
+ cse_var_11: T.int32 = cse_var_2 + 105
+ cse_var_10: T.int32 = cse_var_2 + 104
+ cse_var_9: T.int32 = cse_var_2 + 103
+ cse_var_8: T.int32 = cse_var_2 + 102
+ cse_var_7: T.int32 = cse_var_2 + 101
+ cse_var_6: T.int32 = cse_var_2 + 100
+ cse_var_5: T.int32 = cse_var_2 + 10
+ cse_var_4: T.int32 = cse_var_2 + 1
+ cse_var_3: T.int32 = i0_outer_i1_outer_fused // 16 * 4096 + i_outer_inner * 2048
placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
placeholder_7 = T.Buffer((32768,), data=placeholder.data)
placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
- compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 16 * 8192 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- for i0_inner, i1_inner in T.grid(32, 32):
- cse_var_4: T.int32 = i0_outer_i1_outer_fused // 16 * 16384 + i0_inner * 512 + i0_outer_i1_outer_fused % 16 * 32 + i1_inner
+ compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_63] = compute_2[cse_var_63] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_88] = compute_2[cse_var_88] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_97] = compute_2[cse_var_97] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_106] = compute_2[cse_var_106] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_107] = compute_2[cse_var_107] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_114] = compute_2[cse_var_114] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_125] = compute_2[cse_var_125] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_126] = compute_2[cse_var_126] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_19] = compute_2[cse_var_19] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_22] = compute_2[cse_var_22] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_33] = compute_2[cse_var_33] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_38] = compute_2[cse_var_38] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+ compute_2[cse_var_89] = compute_2[cse_var_89] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_90] = compute_2[cse_var_90] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_91] = compute_2[cse_var_91] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_92] = compute_2[cse_var_92] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_93] = compute_2[cse_var_93] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_94] = compute_2[cse_var_94] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_95] = compute_2[cse_var_95] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_96] = compute_2[cse_var_96] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_98] = compute_2[cse_var_98] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_99] = compute_2[cse_var_99] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_100] = compute_2[cse_var_100] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_101] = compute_2[cse_var_101] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_102] = compute_2[cse_var_102] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_103] = compute_2[cse_var_103] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_104] = compute_2[cse_var_104] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_105] = compute_2[cse_var_105] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+ compute_2[cse_var_108] = compute_2[cse_var_108] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_109] = compute_2[cse_var_109] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_110] = compute_2[cse_var_110] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_111] = compute_2[cse_var_111] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_112] = compute_2[cse_var_112] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_113] = compute_2[cse_var_113] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_115] = compute_2[cse_var_115] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_116] = compute_2[cse_var_116] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_117] = compute_2[cse_var_117] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_118] = compute_2[cse_var_118] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_119] = compute_2[cse_var_119] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_120] = compute_2[cse_var_120] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_121] = compute_2[cse_var_121] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_122] = compute_2[cse_var_122] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_123] = compute_2[cse_var_123] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_124] = compute_2[cse_var_124] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+ compute_2[cse_var_127] = compute_2[cse_var_127] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_128] = compute_2[cse_var_128] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_129] = compute_2[cse_var_129] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_130] = compute_2[cse_var_130] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_18] = compute_2[cse_var_18] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+ compute_2[cse_var_20] = compute_2[cse_var_20] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_21] = compute_2[cse_var_21] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_23] = compute_2[cse_var_23] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_24] = compute_2[cse_var_24] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_25] = compute_2[cse_var_25] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_26] = compute_2[cse_var_26] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_27] = compute_2[cse_var_27] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_28] = compute_2[cse_var_28] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_29] = compute_2[cse_var_29] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_30] = compute_2[cse_var_30] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_31] = compute_2[cse_var_31] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_32] = compute_2[cse_var_32] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_34] = compute_2[cse_var_34] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_35] = compute_2[cse_var_35] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_36] = compute_2[cse_var_36] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_37] = compute_2[cse_var_37] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1024], T.float32(0))
+ compute_2[cse_var_39] = compute_2[cse_var_39] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_40] = compute_2[cse_var_40] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_41] = compute_2[cse_var_41] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_42] = compute_2[cse_var_42] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_43] = compute_2[cse_var_43] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_44] = compute_2[cse_var_44] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_45] = compute_2[cse_var_45] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_46] = compute_2[cse_var_46] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_47] = compute_2[cse_var_47] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_48] = compute_2[cse_var_48] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_49] = compute_2[cse_var_49] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_50] = compute_2[cse_var_50] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_51] = compute_2[cse_var_51] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_52] = compute_2[cse_var_52] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_53] = compute_2[cse_var_53] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_54] = compute_2[cse_var_54] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1280], T.float32(0))
+ compute_2[cse_var_55] = compute_2[cse_var_55] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_56] = compute_2[cse_var_56] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_57] = compute_2[cse_var_57] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_58] = compute_2[cse_var_58] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_59] = compute_2[cse_var_59] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_60] = compute_2[cse_var_60] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_61] = compute_2[cse_var_61] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_62] = compute_2[cse_var_62] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_64] = compute_2[cse_var_64] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_65] = compute_2[cse_var_65] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_66] = compute_2[cse_var_66] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_67] = compute_2[cse_var_67] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_68] = compute_2[cse_var_68] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_69] = compute_2[cse_var_69] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_70] = compute_2[cse_var_70] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_71] = compute_2[cse_var_71] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1536], T.float32(0))
+ compute_2[cse_var_72] = compute_2[cse_var_72] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_73] = compute_2[cse_var_73] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 1] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_74] = compute_2[cse_var_74] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 2] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_75] = compute_2[cse_var_75] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 3] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_76] = compute_2[cse_var_76] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 4] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_77] = compute_2[cse_var_77] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 5] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_78] = compute_2[cse_var_78] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 6] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_79] = compute_2[cse_var_79] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 7] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_80] = compute_2[cse_var_80] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 8] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_81] = compute_2[cse_var_81] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 9] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_82] = compute_2[cse_var_82] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 10] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_83] = compute_2[cse_var_83] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 11] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_84] = compute_2[cse_var_84] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 12] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_85] = compute_2[cse_var_85] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 13] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_86] = compute_2[cse_var_86] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 14] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ compute_2[cse_var_87] = compute_2[cse_var_87] + placeholder_6[placeholder_5[cse_var_1] * 16 + cse_var_131 + 15] * T.max(placeholder_7[cse_var_3 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 1792], T.float32(0))
+ for i0_inner, i1_inner in T.grid(16, 32):
+ cse_var_132: T.int32 = i0_outer_i1_outer_fused // 16 * 8192 + i0_inner * 512 + i0_outer_i1_outer_fused % 16 * 32 + i1_inner
compute_3 = T.Buffer((65536,), data=compute.data)
placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
- compute_3[cse_var_4] = T.max(compute_2[i0_inner * 32 + i1_inner] + placeholder_5[cse_var_4], T.float32(0))
+ compute_3[cse_var_132] = T.max(compute_2[i0_inner * 32 + i1_inner] + placeholder_5[cse_var_132], T.float32(0))
</pre></div>
</div>
</div>
@@ -690,7 +1071,7 @@ class Module:
<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: 1.656 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 2.751 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 055ee8afb2..50a1e2a2af 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<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>00:47.212</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:42.856</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,11 +354,11 @@
</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>00:47.168</p></td>
+<td><p>00:42.818</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>00:00.027</p></td>
+<td><p>00:00.021</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>
@@ -366,11 +366,11 @@
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></td>
-<td><p>00:00.006</p></td>
+<td><p>00:00.005</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_arm.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-arm-py"><span class="std std-ref">Auto-tuning a Convolutional Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_arm.py</span></code>)</p></td>
-<td><p>00:00.006</p></td>
+<td><p>00:00.005</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
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 513453623e..14d1984a54 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -573,7 +573,8 @@ for this template</p>
waiting for device...
device available
Get devices for measurement successfully!
-No: 1 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+No: 1 GFLOPS: 338.11/338.11 result: MeasureResult(costs=(0.0006847002883435583,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0730724334716797, timestamp=1683281099.2884815) [('tile_f', [-1, 1, 32, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5490850
+No: 2 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -695,8 +696,26 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5171664
-No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 32, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3149212
+No: 3 GFLOPS: 0.00/338.11 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
+ return self.__get_result()
+ File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
+ raise self._exception
+ File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
+ result = self.fn(*self.args, **self.kwargs)
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 432, in <lambda>
+ worker = lambda *args: self._worker_run(*args)
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 401, in _worker_run
+ return proc.recv()
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 309, in recv
+ raise TimeoutError()
+TimeoutError
+
+ [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3690178
+No: 4 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -818,8 +837,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8050664
-No: 3 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3076143
+No: 5 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -941,8 +960,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 64, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2898546
-No: 4 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1105001
+No: 6 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1064,10 +1083,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2015182
-No: 5 GFLOPS: 40.55/40.55 result: MeasureResult(costs=(0.005709265590909091,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.552736520767212, timestamp=1683267721.9495256) [('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5906714
-No: 6 GFLOPS: 40.48/40.55 result: MeasureResult(costs=(0.005719022111111111,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.378983497619629, timestamp=1683267722.848646) [('tile_f', [-1, 32, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4656104
-No: 7 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,614915
+No: 7 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1189,8 +1206,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5021600
-No: 8 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3932741
+No: 8 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1312,9 +1329,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 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,10194376
-No: 9 GFLOPS: 7.35/40.55 result: MeasureResult(costs=(0.031504911249999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.7945895195007324, timestamp=1683267726.8854792) [('tile_f', [-1, 4, 2, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9293632
-No: 10 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 128, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,280991
+No: 9 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1436,8 +1452,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 256, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8270339
-No: 11 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 128]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,307110
+No: 10 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1559,8 +1575,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9261023
-No: 12 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 512, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1889589
+No: 11 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1682,8 +1698,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5938384
-No: 13 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 64, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6259485
+No: 12 GFLOPS: 40.04/338.11 result: MeasureResult(costs=(0.005781580037037037,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.5704667568206787, timestamp=1683281109.8871222) [('tile_f', [-1, 8, 16, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8784509
+No: 13 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1805,8 +1822,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4508921
-No: 14 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4510715
+No: 14 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1928,26 +1945,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6801698
-No: 15 GFLOPS: 0.00/40.55 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
- return self.__get_result()
- File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
- raise self._exception
- File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
- result = self.fn(*self.args, **self.kwargs)
- File "/workspace/python/tvm/contrib/popen_pool.py", line 432, in <lambda>
- worker = lambda *args: self._worker_run(*args)
- File "/workspace/python/tvm/contrib/popen_pool.py", line 401, in _worker_run
- return proc.recv()
- File "/workspace/python/tvm/contrib/popen_pool.py", line 309, in recv
- raise TimeoutError()
-TimeoutError
-
- [('tile_f', [-1, 1, 1, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10315496
-No: 16 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 16, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1251448
+No: 15 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2069,8 +2068,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, 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, 256, 1, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,867963
-No: 17 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,697246
+No: 16 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2192,8 +2191,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3906436
-No: 18 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2271861
+No: 17 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2315,8 +2314,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3201082
-No: 19 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 64, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1408708
+No: 18 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2438,8 +2437,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 32, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3960483
-No: 20 GFLOPS: 0.00/40.55 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7541027
+No: 19 GFLOPS: 19.58/338.11 result: MeasureResult(costs=(0.011821615,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.6270549297332764, timestamp=1683281112.9016883) [('tile_f', [-1, 8, 1, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3257323
+No: 20 GFLOPS: 0.00/338.11 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2561,7 +2561,7 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2128011
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4826081
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2600,9 +2600,9 @@ and measure running time.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Finish loading 20 records
Best config:
-[('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5906714
+[('tile_f', [-1, 1, 32, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5490850
Finish loading 20 records
-Time cost of this operator: 0.006087
+Time cost of this operator: 0.001033
</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 6b68659974..e404ae51cb 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -649,10 +649,10 @@ the tuned operator.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 318.8 98.738 (1, 2, 10, 10, 3) 2 1 [318.8]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.108 0.963 (1, 6, 10, 10) 1 1 [3.108]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.965 0.299 (1, 1, 10, 10, 3) 1 1 [0.965]
-Total_time - 322.873 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 316.4 98.752 (1, 2, 10, 10, 3) 2 1 [316.4]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.045 0.95 (1, 6, 10, 10) 1 1 [3.045]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.953 0.298 (1, 1, 10, 10, 3) 1 1 [0.953]
+Total_time - 320.398 - - - - -
</pre></div>
</div>
</div>
@@ -704,13 +704,13 @@ Total_time -
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 103.4 97.521 (1, 6, 10, 10, 1) 2 1 [103.4]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.773 1.672 (1, 6, 10, 10) 1 1 [1.773]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.855 0.807 (1, 3, 10, 10, 1) 1 1 [0.855]
-Total_time - 106.029 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 103.0 97.484 (1, 6, 10, 10, 1) 2 1 [103.0]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.781 1.686 (1, 6, 10, 10) 1 1 [1.781]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.878 0.831 (1, 3, 10, 10, 1) 1 1 [0.878]
+Total_time - 105.659 - - - - -
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 35.988 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 29.610 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/9ccca8fd489a1486ac71b55a55c320c5/micro_autotune.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_autotune.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_pytorch.html b/docs/how_to/work_with_microtvm/micro_pytorch.html
index 8a0c7992b8..44abe3775e 100644
--- a/docs/how_to/work_with_microtvm/micro_pytorch.html
+++ b/docs/how_to/work_with_microtvm/micro_pytorch.html
@@ -460,7 +460,7 @@ download a cat image and preprocess it to use as the model input.</p>
Downloading: "https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
0%| | 0.00/3.42M [00:00<?, ?B/s]
-100%|##########| 3.42M/3.42M [00:00<00:00, 81.5MB/s]
+100%|##########| 3.42M/3.42M [00:00<00:00, 39.7MB/s]
/workspace/python/tvm/relay/frontend/pytorch_utils.py:47: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
return LooseVersion(torch_ver) > ver
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
@@ -586,7 +586,7 @@ via the host <cite>main.cc`</cite> or if a Zephyr emulated board is selected as
Torch top-1 id: 282, class name: tiger cat
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 26.593 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 23.351 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-pytorch-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/12b9ecc04c41abaa12022061771821d1/micro_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">micro_pytorch.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index c058ed59a9..4288b2841d 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -528,7 +528,7 @@ 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/tmpt3mzqdty/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmp9yxg_u8h/images/random'
</pre></div>
</div>
</div>
@@ -588,8 +588,8 @@ 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="[0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [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/tmpt3mzqdty/images/target contains 8144 images
-/tmp/tmpt3mzqdty/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmp9yxg_u8h/images/target contains 8144 images
+/tmp/tmp9yxg_u8h/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -701,13 +701,13 @@ 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 - 43s - loss: 0.2601 - accuracy: 0.9158 - val_loss: 0.1228 - val_accuracy: 0.9562 - 43s/epoch - 132ms/step
+328/328 - 41s - loss: 0.2093 - accuracy: 0.9268 - val_loss: 0.1127 - val_accuracy: 0.9569 - 41s/epoch - 125ms/step
Epoch 2/3
-328/328 - 37s - loss: 0.0967 - accuracy: 0.9618 - val_loss: 0.1026 - val_accuracy: 0.9649 - 37s/epoch - 111ms/step
+328/328 - 35s - loss: 0.1037 - accuracy: 0.9624 - val_loss: 0.0929 - val_accuracy: 0.9653 - 35s/epoch - 108ms/step
Epoch 3/3
-328/328 - 36s - loss: 0.0631 - accuracy: 0.9779 - val_loss: 0.1065 - val_accuracy: 0.9611 - 36s/epoch - 111ms/step
+328/328 - 35s - loss: 0.0718 - accuracy: 0.9727 - val_loss: 0.0874 - val_accuracy: 0.9694 - 35s/epoch - 107ms/step
-<keras.callbacks.History object at 0x7fb2053a1fd0>
+<keras.callbacks.History object at 0x7fa9533a0b50>
</pre></div>
</div>
</div>
@@ -971,7 +971,7 @@ 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> ( 4 minutes 51.119 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 37.968 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 62c5dfeed8..8a3c09f5c3 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<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>08:22.699</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>07:58.155</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 82%" />
@@ -354,27 +354,27 @@
</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">5. 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>04:51.119</p></td>
+<td><p>04:37.968</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">6. Model Tuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>01:35.988</p></td>
+<td><p>01:29.610</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_pytorch.html#sphx-glr-how-to-work-with-microtvm-micro-pytorch-py"><span class="std std-ref">4. microTVM PyTorch Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_pytorch.py</span></code>)</p></td>
-<td><p>01:26.593</p></td>
+<td><p>01:23.351</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_aot.html#sphx-glr-how-to-work-with-microtvm-micro-aot-py"><span class="std std-ref">3. microTVM Ahead-of-Time (AOT) Compilation</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_aot.py</span></code>)</p></td>
-<td><p>00:11.533</p></td>
+<td><p>00:10.862</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_custom_ide.html#sphx-glr-how-to-work-with-microtvm-micro-custom-ide-py"><span class="std std-ref">9. Bring microTVM to your own development environment</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_custom_ide.py</span></code>)</p></td>
-<td><p>00:09.159</p></td>
+<td><p>00:08.646</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><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">2. microTVM TFLite Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:08.307</p></td>
+<td><p>00:07.718</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><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">7. 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 2484fe16ad..e2ceff30dc 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -345,7 +345,7 @@
<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>00:42.141</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:38.817</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,19 +354,19 @@
</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>00:36.987</p></td>
+<td><p>00:34.016</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>00:03.312</p></td>
+<td><p>00:03.093</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>00:01.834</p></td>
+<td><p>00:01.702</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>
-<td><p>00:00.008</p></td>
+<td><p>00:00.007</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/work_with_schedules/intrin_math.html b/docs/how_to/work_with_schedules/intrin_math.html
index 13d4e93b7d..03a432cb54 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -554,7 +554,7 @@ 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 0x7fae526ed950>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7fa59e48a950>
</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 e82bbaacc8..40cc105d91 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -345,7 +345,7 @@
<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>00:09.573</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:07.020</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,35 +354,35 @@
</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>00:06.494</p></td>
+<td><p>00:04.149</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>00:01.375</p></td>
+<td><p>00:01.268</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>00:00.703</p></td>
+<td><p>00:00.664</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>00:00.672</p></td>
+<td><p>00:00.648</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>00:00.153</p></td>
+<td><p>00:00.137</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><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>00:00.073</p></td>
+<td><p>00:00.065</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><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>00:00.067</p></td>
+<td><p>00:00.058</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>00:00.036</p></td>
+<td><p>00:00.031</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/install/nnpack.html b/docs/install/nnpack.html
index 60011897a9..2412542d0d 100644
--- a/docs/install/nnpack.html
+++ b/docs/install/nnpack.html
@@ -234,17 +234,7 @@
<p class="caption" role="heading"><span class="caption-text">Getting Started</span></p>
<ul class="current">
<li class="toctree-l1 current"><a class="reference internal" href="index.html">Installing TVM</a><ul class="current">
-<li class="toctree-l2 current"><a class="reference internal" href="from_source.html">Install from Source</a><ul class="current">
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#developers-get-source-from-github">Developers: Get Source from Github</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#build-the-shared-library">Build the Shared Library</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#python-package-installation">Python Package Installation</a></li>
-<li class="toctree-l3 current"><a class="reference internal" href="from_source.html#install-contrib-libraries">Install Contrib Libraries</a><ul class="current">
-<li class="toctree-l4 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a></li>
-</ul>
-</li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#enable-c-tests">Enable C++ Tests</a></li>
-</ul>
-</li>
+<li class="toctree-l2"><a class="reference internal" href="from_source.html">Install from Source</a></li>
<li class="toctree-l2"><a class="reference internal" href="docker.html">Docker Images</a></li>
<li class="toctree-l2 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#conditions">Conditions</a></li>
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index d0562e09e0..670a8b26df 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1622,7 +1622,7 @@ 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>
@@ -1906,7 +1906,7 @@ 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 aa3f36c5c9..4a1b7b516b 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/706a5b2dd/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L45">rpc_server.ts:45</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/706a5b2dd/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L45">rpc_server.ts:45</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/706a5b2dd/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L44">rpc_server.ts:44</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/706a5b2dd/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L65">rpc_server.ts:65</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/706a5b2dd/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L51">rpc_server.ts:51</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/706a5b2dd/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L59">rpc_server.ts:59</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 e8df98a50c..312939b454 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/706a5b2dd/web/src/memory.ts#L223">memory.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L208">memory.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L312">memory.ts:312</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L284">memory.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L388">memory.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L376">memory.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L267">memory.ts:267</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L243">memory.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L321">memory.ts:321</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L252">memory.ts:252</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L359">memory.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/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/706a5b2dd/web/src/memory.ts#L342">memory.ts:342</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L342">memory.ts:342</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -496,7 +496,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L350">memory.ts:350</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L350">memory.ts:350</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -522,7 +522,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L326">memory.ts:326</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L326">memory.ts:326</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -548,7 +548,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L363">memory.ts:363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L363">memory.ts:363</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -574,7 +574,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L346">memory.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L346">memory.ts:346</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -600,7 +600,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L334">memory.ts:334</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L334">memory.ts:334</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/dldatatype.html b/docs/reference/api/typedoc/classes/dldatatype.html
index 44485c7f2e..70df987c46 100644
--- a/docs/reference/api/typedoc/classes/dldatatype.html
+++ b/docs/reference/api/typedoc/classes/dldatatype.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/706a5b2dd/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L359">runtime.ts:359</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
<div class="tsd-signature tsd-kind-icon">bits<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L357">runtime.ts:357</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L357">runtime.ts:357</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">code<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L355">runtime.ts:355</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -177,7 +177,7 @@
<div class="tsd-signature tsd-kind-icon">lanes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L359">runtime.ts:359</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -199,7 +199,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L376">runtime.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L376">runtime.ts:376</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -216,7 +216,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L367">runtime.ts:367</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L367">runtime.ts:367</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/dldevice.html b/docs/reference/api/typedoc/classes/dldevice.html
index 6001fdbfb8..7effdef3c0 100644
--- a/docs/reference/api/typedoc/classes/dldevice.html
+++ b/docs/reference/api/typedoc/classes/dldevice.html
@@ -118,7 +118,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L299">runtime.ts:299</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L299">runtime.ts:299</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L297">runtime.ts:297</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -161,7 +161,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L295">runtime.ts:295</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -183,7 +183,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L320">runtime.ts:320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L320">runtime.ts:320</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -205,7 +205,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L327">runtime.ts:327</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L327">runtime.ts:327</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/environment.html b/docs/reference/api/typedoc/classes/environment.html
index b482049ea5..07bc05e489 100644
--- a/docs/reference/api/typedoc/classes/environment.html
+++ b/docs/reference/api/typedoc/classes/environment.html
@@ -125,7 +125,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/environment.ts#L86">environment.ts:86</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/environment.ts#L86">environment.ts:86</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -169,7 +169,7 @@
<aside class="tsd-sources">
<p>Implementation of <a href="../interfaces/libraryprovider.html">LibraryProvider</a>.<a href="../interfaces/libraryprovider.html#imports">imports</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/environment.ts#L70">environment.ts:70</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/environment.ts#L70">environment.ts:70</a></li>
</ul>
</aside>
</section>
@@ -179,7 +179,7 @@
<div class="tsd-signature tsd-kind-icon">logger<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>msg<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/environment.ts#L69">environment.ts:69</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/environment.ts#L69">environment.ts:69</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -210,7 +210,7 @@
<div class="tsd-signature tsd-kind-icon">packedCFunc<wbr>Table<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">ctypes.FTVMWasmPackedCFunc</span><span class="tsd-signature-symbol"> | </span><span class="tsd-signature-type">undefined</span><span class="tsd-signature-symbol">></span><span class="tsd-signature-symbol"> = [undefined,]</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/environment.ts#L78">environment.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/environment.ts#L78">environment.ts:78</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -228,7 +228,7 @@
<div class="tsd-signature tsd-kind-icon">packedCFunc<wbr>Table<wbr>Free<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">number</span><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/706a5b2dd/web/src/environment.ts#L84">environment.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/environment.ts#L84">environment.ts:84</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -250,7 +250,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/environment.ts#L105">environment.ts:105</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/environment.ts#L105">environment.ts:105</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ffilibrary.html b/docs/reference/api/typedoc/classes/ffilibrary.html
index 4888c76858..9bfba4901d 100644
--- a/docs/reference/api/typedoc/classes/ffilibrary.html
+++ b/docs/reference/api/typedoc/classes/ffilibrary.html
@@ -131,7 +131,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L50">runtime.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L50">runtime.ts:50</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -156,7 +156,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L47">runtime.ts:47</a></li>
</ul>
</aside>
</section>
@@ -166,7 +166,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L46">runtime.ts:46</a></li>
</ul>
</aside>
</section>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L45">runtime.ts:45</a></li>
</ul>
</aside>
</section>
@@ -186,7 +186,7 @@
<div class="tsd-signature tsd-kind-icon">webGPUContext<span class="tsd-signature-symbol">:</span> <a href="webgpucontext.html" class="tsd-signature-type">WebGPUContext</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L48">runtime.ts:48</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L48">runtime.ts:48</a></li>
</ul>
</aside>
</section>
@@ -203,7 +203,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L77">runtime.ts:77</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L77">runtime.ts:77</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -226,7 +226,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L67">runtime.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L67">runtime.ts:67</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -243,7 +243,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L85">runtime.ts:85</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L85">runtime.ts:85</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <a href="cachedcallstack.html" class="tsd-signature-type">CachedCallStack</a></h4>
@@ -260,7 +260,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L96">runtime.ts:96</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L96">runtime.ts:96</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -283,7 +283,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L73">runtime.ts:73</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L73">runtime.ts:73</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/instance.html b/docs/reference/api/typedoc/classes/instance.html
index b3b4c0ed6e..4cfe5ff7ce 100644
--- a/docs/reference/api/typedoc/classes/instance.html
+++ b/docs/reference/api/typedoc/classes/instance.html
@@ -161,7 +161,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L844">runtime.ts:844</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L844">runtime.ts:844</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L834">runtime.ts:834</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L834">runtime.ts:834</a></li>
</ul>
</aside>
</section>
@@ -234,7 +234,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L833">runtime.ts:833</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L833">runtime.ts:833</a></li>
</ul>
</aside>
</section>
@@ -251,7 +251,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L973">runtime.ts:973</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L973">runtime.ts:973</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -296,7 +296,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L932">runtime.ts:932</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -318,7 +318,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L901">runtime.ts:901</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L901">runtime.ts:901</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -381,7 +381,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -412,7 +412,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -453,7 +453,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -491,7 +491,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L922">runtime.ts:922</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L922">runtime.ts:922</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -508,7 +508,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -552,7 +552,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L943">runtime.ts:943</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L943">runtime.ts:943</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -577,7 +577,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -609,7 +609,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -640,7 +640,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -672,7 +672,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -695,7 +695,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -729,7 +729,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L986">runtime.ts:986</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L986">runtime.ts:986</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -769,7 +769,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -817,7 +817,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -857,7 +857,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -900,7 +900,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -938,7 +938,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -990,7 +990,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1014,7 +1014,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1046,7 +1046,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1078,7 +1078,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1110,7 +1110,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1141,7 +1141,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L957">runtime.ts:957</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L957">runtime.ts:957</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/memory.html b/docs/reference/api/typedoc/classes/memory.html
index 0118afdb9f..17c0988736 100644
--- a/docs/reference/api/typedoc/classes/memory.html
+++ b/docs/reference/api/typedoc/classes/memory.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L40">memory.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L40">memory.ts:40</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -152,7 +152,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Memory</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L32">memory.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L32">memory.ts:32</a></li>
</ul>
</aside>
</section>
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span><span class="tsd-signature-symbol"> = true</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L33">memory.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L33">memory.ts:33</a></li>
</ul>
</aside>
</section>
@@ -179,7 +179,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L154">memory.ts:154</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L154">memory.ts:154</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -210,7 +210,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L90">memory.ts:90</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L90">memory.ts:90</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -233,7 +233,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L97">memory.ts:97</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L97">memory.ts:97</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -256,7 +256,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L74">memory.ts:74</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L74">memory.ts:74</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -279,7 +279,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L81">memory.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L81">memory.ts:81</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -302,7 +302,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L104">memory.ts:104</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L104">memory.ts:104</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -325,7 +325,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L132">memory.ts:132</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L132">memory.ts:132</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -362,7 +362,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L145">memory.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L145">memory.ts:145</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -393,7 +393,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L60">memory.ts:60</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L60">memory.ts:60</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -416,7 +416,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L67">memory.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L67">memory.ts:67</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -439,7 +439,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L53">memory.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L53">memory.ts:53</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -462,7 +462,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L114">memory.ts:114</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L114">memory.ts:114</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -485,7 +485,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L124">memory.ts:124</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L124">memory.ts:124</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -502,7 +502,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/memory.ts#L175">memory.ts:175</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/memory.ts#L175">memory.ts:175</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/module.html b/docs/reference/api/typedoc/classes/module.html
index 0358905d2b..f12fc90d8b 100644
--- a/docs/reference/api/typedoc/classes/module.html
+++ b/docs/reference/api/typedoc/classes/module.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/706a5b2dd/web/src/runtime.ts#L614">runtime.ts:614</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L614">runtime.ts:614</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -169,7 +169,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L626">runtime.ts:626</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L626">runtime.ts:626</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -186,7 +186,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L653">runtime.ts:653</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L653">runtime.ts:653</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -218,7 +218,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L641">runtime.ts:641</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L641">runtime.ts:641</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -250,7 +250,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L687">runtime.ts:687</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L687">runtime.ts:687</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ndarray.html b/docs/reference/api/typedoc/classes/ndarray.html
index 7206178f5a..476b0bad81 100644
--- a/docs/reference/api/typedoc/classes/ndarray.html
+++ b/docs/reference/api/typedoc/classes/ndarray.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L401">runtime.ts:401</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L401">runtime.ts:401</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <a href="dldevice.html" class="tsd-signature-type">DLDevice</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L394">runtime.ts:394</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L394">runtime.ts:394</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -173,7 +173,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L390">runtime.ts:390</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L390">runtime.ts:390</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -188,7 +188,7 @@
<div class="tsd-signature tsd-kind-icon">ndim<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L388">runtime.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L388">runtime.ts:388</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -203,7 +203,7 @@
<div class="tsd-signature tsd-kind-icon">shape<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L392">runtime.ts:392</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L392">runtime.ts:392</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -225,7 +225,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L480">runtime.ts:480</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L480">runtime.ts:480</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -258,7 +258,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L524">runtime.ts:524</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L524">runtime.ts:524</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -290,7 +290,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L465">runtime.ts:465</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L465">runtime.ts:465</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -307,7 +307,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L458">runtime.ts:458</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L458">runtime.ts:458</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -339,7 +339,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L584">runtime.ts:584</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L584">runtime.ts:584</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -363,7 +363,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L553">runtime.ts:553</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L553">runtime.ts:553</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/packedfunccell.html b/docs/reference/api/typedoc/classes/packedfunccell.html
index 70848ca6db..d52c592bf9 100644
--- a/docs/reference/api/typedoc/classes/packedfunccell.html
+++ b/docs/reference/api/typedoc/classes/packedfunccell.html
@@ -117,7 +117,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L248">runtime.ts:248</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L248">runtime.ts:248</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L255">runtime.ts:255</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L255">runtime.ts:255</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -163,7 +163,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L264">runtime.ts:264</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L264">runtime.ts:264</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/rpcserver.html b/docs/reference/api/typedoc/classes/rpcserver.html
index 96e7cdfc73..3073d0eda3 100644
--- a/docs/reference/api/typedoc/classes/rpcserver.html
+++ b/docs/reference/api/typedoc/classes/rpcserver.html
@@ -115,7 +115,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">get<wbr>Imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">unknown</span><span class="tsd-signat [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -201,7 +201,7 @@
<div class="tsd-signature tsd-kind-icon">key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
</ul>
</aside>
</section>
@@ -211,7 +211,7 @@
<div class="tsd-signature tsd-kind-icon">logger<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>msg<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -242,7 +242,7 @@
<div class="tsd-signature tsd-kind-icon">socket<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">WebSocket</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
</ul>
</aside>
</section>
@@ -252,7 +252,7 @@
<div class="tsd-signature tsd-kind-icon">state<span class="tsd-signature-symbol">:</span> <a href="../enums/rpcserverstate.html" class="tsd-signature-type">RPCServerState</a><span class="tsd-signature-symbol"> = RPCServerState.InitHeader</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
</ul>
</aside>
</section>
@@ -262,7 +262,7 @@
<div class="tsd-signature tsd-kind-icon">url<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/classes/runtimecontext.html b/docs/reference/api/typedoc/classes/runtimecontext.html
index 3f1e18ebc1..3e967767e2 100644
--- a/docs/reference/api/typedoc/classes/runtimecontext.html
+++ b/docs/reference/api/typedoc/classes/runtimecontext.html
@@ -132,7 +132,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L148">runtime.ts:148</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L148">runtime.ts:148</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -172,7 +172,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Item<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L143">runtime.ts:143</a></li>
</ul>
</aside>
</section>
@@ -182,7 +182,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Size<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L144">runtime.ts:144</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L144">runtime.ts:144</a></li>
</ul>
</aside>
</section>
@@ -192,7 +192,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Make<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L145">runtime.ts:145</a></li>
</ul>
</aside>
</section>
@@ -202,7 +202,7 @@
<div class="tsd-signature tsd-kind-icon">get<wbr>Sys<wbr>Lib<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L146">runtime.ts:146</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L146">runtime.ts:146</a></li>
</ul>
</aside>
</section>
@@ -219,7 +219,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L189">runtime.ts:189</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L189">runtime.ts:189</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -263,7 +263,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L163">runtime.ts:163</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L163">runtime.ts:163</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -280,7 +280,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L208">runtime.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L208">runtime.ts:208</a></li>
</ul>
</aside>
<h4 class="tsd-type-parameters-title">Type parameters</h4>
@@ -309,7 +309,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L157">runtime.ts:157</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -326,7 +326,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L167">runtime.ts:167</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L167">runtime.ts:167</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -343,7 +343,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L198">runtime.ts:198</a></li>
</ul>
</aside>
<h4 class="tsd-type-parameters-title">Type parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/scalar.html b/docs/reference/api/typedoc/classes/scalar.html
index f8d34e8962..5052a9117e 100644
--- a/docs/reference/api/typedoc/classes/scalar.html
+++ b/docs/reference/api/typedoc/classes/scalar.html
@@ -112,7 +112,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L235">runtime.ts:235</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -137,7 +137,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L235">runtime.ts:235</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -152,7 +152,7 @@
<div class="tsd-signature tsd-kind-icon">value<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L233">runtime.ts:233</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L233">runtime.ts:233</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmarray.html b/docs/reference/api/typedoc/classes/tvmarray.html
index fba40d5f56..24534328aa 100644
--- a/docs/reference/api/typedoc/classes/tvmarray.html
+++ b/docs/reference/api/typedoc/classes/tvmarray.html
@@ -133,7 +133,7 @@
<aside class="tsd-sources">
<p>Overrides <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#constructor">constructor</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L784">runtime.ts:784</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L784">runtime.ts:784</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -162,7 +162,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#ctx">ctx</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
</section>
@@ -180,7 +180,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#dispose">dispose</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L715">runtime.ts:715</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -197,7 +197,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L804">runtime.ts:804</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L804">runtime.ts:804</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -230,7 +230,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#gethandle">getHandle</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L730">runtime.ts:730</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/706a5b2dd/web/src/runtime.ts#L796">runtime.ts:796</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L796">runtime.ts:796</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -283,7 +283,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typeindex">typeIndex</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L738">runtime.ts:738</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -306,7 +306,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typekey">typeKey</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L758">runtime.ts:758</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmobject.html b/docs/reference/api/typedoc/classes/tvmobject.html
index e7fb37eaaa..de7f0621f6 100644
--- a/docs/reference/api/typedoc/classes/tvmobject.html
+++ b/docs/reference/api/typedoc/classes/tvmobject.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
<div class="tsd-signature tsd-kind-icon">ctx<span class="tsd-signature-symbol">:</span> <a href="runtimecontext.html" class="tsd-signature-type">RuntimeContext</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
</section>
@@ -175,7 +175,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L715">runtime.ts:715</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -192,7 +192,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L730">runtime.ts:730</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L738">runtime.ts:738</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -246,7 +246,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/runtime.ts#L758">runtime.ts:758</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/webgpucontext.html b/docs/reference/api/typedoc/classes/webgpucontext.html
index a20e57f96c..e7bc95f3f4 100644
--- a/docs/reference/api/typedoc/classes/webgpucontext.html
+++ b/docs/reference/api/typedoc/classes/webgpucontext.html
@@ -120,7 +120,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -145,7 +145,7 @@
<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">GPUDevice</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
</ul>
</aside>
</section>
@@ -155,7 +155,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
</ul>
</aside>
</section>
@@ -172,7 +172,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -209,7 +209,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/706a5b2dd/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/eca6edf7f/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/enums/argtypecode.html b/docs/reference/api/typedoc/enums/argtypecode.html
index 1a0d94ea85..0e77cf6c34 100644
--- a/docs/reference/api/typedoc/enums/argtypecode.html
+++ b/docs/reference/api/typedoc/enums/argtypecode.html
@@ -106,7 +106,7 @@
<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 6</span></div>
<aside class="tsd-sources">
... 1732 lines suppressed ...