You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2022/06/24 19:16:49 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@2c3b56c9d3fa07407eb6dda7f2142f6ba3748ecf)
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 4a2a8fa8b deploying docs (apache/tvm@2c3b56c9d3fa07407eb6dda7f2142f6ba3748ecf)
4a2a8fa8b is described below
commit 4a2a8fa8b8fd49242fc03b36833ed0d2ada9639c
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Fri Jun 24 19:16:44 2022 +0000
deploying docs (apache/tvm@2c3b56c9d3fa07407eb6dda7f2142f6ba3748ecf)
---
.../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 | 2 +-
.../how_to/compile_models/from_pytorch.rst.txt | 2 +-
.../how_to/compile_models/from_tensorflow.rst.txt | 2 +-
.../compile_models/sg_execution_times.rst.txt | 22 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 16 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 4 +-
.../how_to/extend_tvm/sg_execution_times.rst.txt | 8 +-
.../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 | 3035 ++++++++------------
.../tune_network_cuda.rst.txt | 2 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 111 +-
.../tune_with_autotvm/sg_execution_times.rst.txt | 10 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 34 +-
.../work_with_microtvm/micro_autotune.rst.txt | 16 +-
.../how_to/work_with_microtvm/micro_train.rst.txt | 16 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 8 +-
.../work_with_relay/sg_execution_times.rst.txt | 6 +-
.../how_to/work_with_schedules/intrin_math.rst.txt | 2 +-
.../work_with_schedules/sg_execution_times.rst.txt | 18 +-
.../how_to/work_with_schedules/tensorize.rst.txt | 2 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 4 +-
.../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 | 2 +-
docs/_sources/tutorial/autotvm_matmul_x86.rst.txt | 20 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 58 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 22 +-
.../tutorial/tensor_expr_get_started.rst.txt | 42 +-
docs/commit_hash | 2 +-
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_oneflow.html | 74 +-
docs/how_to/compile_models/from_paddle.html | 2 +-
docs/how_to/compile_models/from_pytorch.html | 10 +-
docs/how_to/compile_models/from_tensorflow.html | 2 +-
docs/how_to/compile_models/sg_execution_times.html | 26 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 26 +-
docs/how_to/deploy_models/deploy_prequantized.html | 13 +-
.../deploy_models/deploy_prequantized_tflite.html | 4 +-
docs/how_to/deploy_models/deploy_quantized.html | 2 +-
docs/how_to/deploy_models/deploy_ssd_gluoncv.html | 37 +-
docs/how_to/deploy_models/sg_execution_times.html | 16 +-
.../extend_tvm/bring_your_own_datatypes.html | 4 +-
docs/how_to/extend_tvm/sg_execution_times.html | 8 +-
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 | 3035 ++++++++------------
.../tune_with_autoscheduler/tune_network_cuda.html | 2 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 111 +-
.../tune_with_autotvm/sg_execution_times.html | 14 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 34 +-
docs/how_to/work_with_microtvm/micro_autotune.html | 16 +-
docs/how_to/work_with_microtvm/micro_train.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 8 +-
.../how_to/work_with_relay/sg_execution_times.html | 6 +-
docs/how_to/work_with_schedules/intrin_math.html | 2 +-
.../work_with_schedules/sg_execution_times.html | 18 +-
docs/how_to/work_with_schedules/tensorize.html | 2 +-
docs/reference/api/python/auto_scheduler.html | 4 +-
.../api/typedoc/classes/bytestreamreader.html | 12 +-
.../api/typedoc/classes/cachedcallstack.html | 34 +-
docs/reference/api/typedoc/classes/dldatatype.html | 12 +-
docs/reference/api/typedoc/classes/dldevice.html | 10 +-
.../reference/api/typedoc/classes/environment.html | 12 +-
docs/reference/api/typedoc/classes/ffilibrary.html | 20 +-
.../api/typedoc/classes/graphexecutor.html | 16 +-
docs/reference/api/typedoc/classes/instance.html | 40 +-
docs/reference/api/typedoc/classes/memory.html | 34 +-
docs/reference/api/typedoc/classes/module.html | 10 +-
docs/reference/api/typedoc/classes/ndarray.html | 22 +-
.../api/typedoc/classes/packedfunccell.html | 6 +-
docs/reference/api/typedoc/classes/rpcserver.html | 14 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
.../api/typedoc/classes/webgpucontext.html | 12 +-
docs/reference/api/typedoc/enums/argtypecode.html | 30 +-
.../api/typedoc/enums/aynccallbackcode.html | 4 +-
.../api/typedoc/enums/dldatatypecode.html | 8 +-
.../api/typedoc/enums/rpcserverstate.html | 12 +-
docs/reference/api/typedoc/enums/sizeof.html | 18 +-
docs/reference/api/typedoc/index.html | 112 +-
.../api/typedoc/interfaces/disposable.html | 2 +-
.../api/typedoc/interfaces/functioninfo.html | 6 +-
.../api/typedoc/interfaces/libraryprovider.html | 4 +-
docs/searchindex.js | 2 +-
.../vta/tutorials/autotvm/sg_execution_times.html | 4 +-
.../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 | 2 +-
docs/tutorial/autotvm_matmul_x86.html | 20 +-
docs/tutorial/autotvm_relay_x86.html | 262 +-
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 | 42 +-
121 files changed, 3584 insertions(+), 4390 deletions(-)
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 a101dd87b..73667128a 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -114,7 +114,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip7711d7c2-bb2d-4ab9-b495-a12f8fe83ee8 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip5651072b-8a77-4627-9e37-9128756bd30b 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 f51d5966e..8217547ea 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -112,7 +112,7 @@ 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]
0%| | 16.0k/41.5M [00:00<07:42, 94.0kB/s]
0%| | 48.0k/41.5M [00:00<04:52, 149kB/s]
0%| | 96.0k/41.5M [00:00<03:27, 209kB/s]
0%| | 208k/41.5M [00:00<01:52, 385kB/s]
1%|1 | 432k/41.5M [00:00<00:59, 718kB/s]
2%|2 | 872k/41.5M [00:01<00:31, 1.35MB/s]
4%|4 | 1.72M/41.5M [00:01<00:16, 2.60MB/s]
8%|7 | 3.26M/41.5M [00:01<00:08, 4.70MB/s]
12%|#1 | 4.80M/41.5M [00:01<00:06, 6.11MB/s]
15%|#5 | 6.34M/41.5M [00:01<00:05, 7.07MB/s]
19%|#8 | 7.88M/41.5M [00:01<00:04, 7.72MB/s]
23%|##2 | 9.41M/41.5M [00:02<00:04, 8.17MB/s]
26%|##6 | 10.9M/41.5M [00:02<00:03, 8.47MB/s]
30%|### | 12.5M/41.5M [00:02<00:03, 8.69MB/s]
34%|###3 | 14.0M/41.5M [00:02<00:03, 8.82MB/s]
37%|###7 | 15.5M/41.5M [00:02<00:03, 8.93MB/s]
41%|####1 | 17.1M/41.5M [00:02<0
0:02, 9.02MB/s]
45%|####4 | 18.6M/41.5M [00:03<00:02, 9.06MB/s]
49%|####8 | 20.2M/41.5M [00:03<00:02, 9.10MB/s]
52%|#####2 | 21.7M/41.5M [00:03<00:02, 9.09MB/s]
56%|#####5 | 23.2M/41.5M [00:03<00:02, 9.12MB/s]
60%|#####9 | 24.8M/41.5M [00:03<00:01, 9.15MB/s]
63%|######3 | 26.3M/41.5M [00:04<00:01, 9.16MB/s]
66%|######5 | 27.2M/41.5M [00:04<00:01, 7.56MB/s]
69%|######9 | 28.8M/41.5M [00:04<00:01, 8.03MB/s]
71%|#######1 | 29.6M/41.5M [00:04<00:01, 7.69MB/s]
75%|#######5 | 31.2M/41.5M [00:04<00:01, 8.16MB/s]
79%|#######8 | 32.7M/41.5M [00:04<00:01, 8.47MB/s]
83%|########2 | 34.2M/41.5M [00:05<00:00, 8.70MB/s]
86%|########6 | 35.8M/41.5M [00:05<00:00, 8.84MB/s]
90%|########9 | 37.3M/41.5M [00:05<00:00, 8.95MB/s]
94%|#########3| 38.9M/41.5M [00:05<00:00, 9.00MB/s]
97%|#########7| 40.4M/41.5M [00:05<00:00, 9.06MB/s]
99%|#########9| 41.3M/41.5M [00:06<00:00, 7.35MB/s]
100%|###
#######| 41.5M/41.5M [00:06<00:00, 7.24MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
0%| | 16.0k/41.5M [00:00<08:06, 89.4kB/s]
0%| | 40.0k/41.5M [00:00<06:16, 115kB/s]
0%| | 96.0k/41.5M [00:00<03:31, 205kB/s]
0%| | 168k/41.5M [00:00<02:33, 282kB/s]
1%| | 352k/41.5M [00:00<01:18, 549kB/s]
2%|1 | 664k/41.5M [00:01<00:44, 952kB/s]
3%|3 | 1.30M/41.5M [00:01<00:22, 1.85MB/s]
6%|6 | 2.59M/41.5M [00:01<00:11, 3.61MB/s]
10%|9 | 4.06M/41.5M [00:01<00:07, 5.09MB/s]
13%|#3 | 5.53M/41.5M [00:01<00:06, 6.10MB/s]
17%|#6 | 7.00M/41.5M [00:02<00:05, 6.79MB/s]
20%|## | 8.46M/41.5M [00:02<00:04, 7.26MB/s]
24%|##3 | 9.93M/41.5M [00:02<00:04, 7.59MB/s]
27%|##7 | 11.4M/41.5M [00:02<00:04, 7.84MB/s]
31%|###1 | 12.9M/41.5M [00:02<00:03, 7.99MB/s]
35%|###4 | 14.3M/41.5M [00:02<00:03, 8.09MB/s]
38%|###8 | 15.8M/41.5M [00:03<00
:03, 8.17MB/s]
42%|####1 | 17.3M/41.5M [00:03<00:03, 8.23MB/s]
45%|####5 | 18.7M/41.5M [00:03<00:02, 8.27MB/s]
49%|####8 | 20.2M/41.5M [00:03<00:02, 8.30MB/s]
52%|#####2 | 21.7M/41.5M [00:03<00:02, 8.31MB/s]
56%|#####5 | 23.1M/41.5M [00:04<00:02, 8.33MB/s]
59%|#####9 | 24.6M/41.5M [00:04<00:02, 8.34MB/s]
63%|######2 | 26.1M/41.5M [00:04<00:01, 8.33MB/s]
66%|######6 | 27.5M/41.5M [00:04<00:01, 9.17MB/s]
69%|######9 | 28.8M/41.5M [00:04<00:01, 9.97MB/s]
72%|#######1 | 29.8M/41.5M [00:04<00:01, 8.98MB/s]
74%|#######4 | 30.7M/41.5M [00:04<00:01, 7.84MB/s]
77%|#######6 | 31.9M/41.5M [00:05<00:01, 8.49MB/s]
80%|#######9 | 33.1M/41.5M [00:05<00:00, 9.36MB/s]
82%|########2 | 34.1M/41.5M [00:05<00:00, 8.39MB/s]
84%|########4 | 34.9M/41.5M [00:05<00:00, 7.31MB/s]
88%|########7 | 36.3M/41.5M [00:05<00:00, 8.48MB/s]
90%|######### | 37.5M/41.5M [00:05<00:00, 9.28MB/s]
93%|####
#####2| 38.4M/41.5M [00:05<00:00, 8.31MB/s]
95%|#########4| 39.3M/41.5M [00:06<00:00, 8.19MB/s]
97%|#########7| 40.4M/41.5M [00:06<00:00, 9.07MB/s]
100%|#########9| 41.3M/41.5M [00:06<00:00, 8.04MB/s]
100%|##########| 41.5M/41.5M [00:06<00:00, 6.93MB/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 fa9a431c9..95bd9586b 100644
--- a/docs/_sources/how_to/compile_models/from_paddle.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
@@ -235,7 +235,7 @@ Look up prediction top 1 index in 1000 class synset.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 5.825 seconds)
+ **Total running time of the script:** ( 1 minutes 6.676 seconds)
.. _sphx_glr_download_how_to_compile_models_from_paddle.py:
diff --git a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
index 822c8dd5a..7bbbfb876 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -93,7 +93,7 @@ Load a pretrained PyTorch model
.. code-block:: none
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
0%| | 0.00/44.7M [00:00<?, ?B/s]
9%|8 | 3.95M/44.7M [00:00<00:01, 41.4MB/s]
27%|##7 | 12.2M/44.7M [00:00<00:00, 68.1MB/s]
62%|######2 | 27.8M/44.7M [00:00<00:00, 111MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 122MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
5%|4 | 2.07M/44.7M [00:00<00:02, 21.6MB/s]
13%|#3 | 5.88M/44.7M [00:00<00:01, 32.2MB/s]
29%|##9 | 13.0M/44.7M [00:00<00:00, 51.7MB/s]
59%|#####8 | 26.3M/44.7M [00:00<00:00, 85.9MB/s]
82%|########1 | 36.4M/44.7M [00:00<00:00, 93.3MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 79.5MB/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 be5f4b5f8..56d79195b 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -422,7 +422,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 1.946 seconds)
+ **Total running time of the script:** ( 1 minutes 1.995 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 ae7904548..b09b872ae 100644
--- a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
@@ -5,26 +5,26 @@
Computation times
=================
-**05:22.379** total execution time for **how_to_compile_models** files:
+**05:40.658** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 01:05.825 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 01:06.676 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:01.946 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:01.995 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 00:57.694 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 00:59.175 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:31.859 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:39.645 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:23.901 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:32.333 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:23.763 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:23.262 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:21.567 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:21.589 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:19.344 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:19.486 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:14.126 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:14.102 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.355 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.395 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
index 20c703230..0fbbcaa67 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
@@ -440,7 +440,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.9979 15.9347 16.5312 15.7462 0.2189
+ 16.0560 16.0311 16.3316 15.9755 0.0957
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 765e0cd0a..a99a8f252 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
@@ -122,7 +122,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
.. code-block:: none
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
0%| | 0.00/170M [00:00<?, ?B/s]
1%| | 1.27M/170M [00:00<00:13, 13.1MB/s]
2%|2 | 3.42M/170M [00:00<00:09, 18.4MB/s]
5%|4 | 8.40M/170M [00:00<00:05, 33.6MB/s]
11%|#1 | 19.1M/170M [00:00<00:02, 64.2MB/s]
24%|##3 | 40.3M/170M [00:00<00:01, 121MB/s]
37%|###7 | 63.6M/170M [00:00<00:00, 163MB/s]
51%|#####1 | 87.3M/170M [00:00<00:00, 191MB/s]
65%|######5 | 111M/170M [00:00<00:00, 208MB/s]
79%|#######9 | 135M/170M [00:00<00:00, 221MB/s]
93%|#########3| 158M/170M [00:01<00:00, 229MB/s]
100%|##########| 170M/170M [00:01<00:00, 169MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
1%| | 944k/170M [00:00<00:18, 9.59MB/s]
2%|1 | 2.64M/170M [00:00<00:12, 14.5MB/s]
3%|3 | 5.83M/170M [00:00<00:07, 23.0MB/s]
7%|6 | 11.7M/170M [00:00<00:04, 38.2MB/s]
14%|#3 | 23.0M/170M [00:00<00:02, 66.8MB/s]
22%|##1 | 37.1M/170M [00:00<00:01, 94.6MB/s]
33%|###3 | 56.4M/170M [00:00<00:00, 130MB/s]
45%|####4 | 76.2M/170M [00:00<00:00, 154MB/s]
56%|#####6 | 95.7M/170M [00:00<00:00, 170MB/s]
68%|######7 | 115M/170M [00:01<00:00, 181MB/s]
80%|#######9 | 135M/170M [00:01<00:00, 189MB/s]
91%|#########1| 155M/170M [00:01<00:00, 194MB/s]
100%|##########| 170M/170M [00:01<00:00, 138MB/s]
/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:3878: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/usr/local/lib/python3.7/dist-packages/torchvision/models/detection/anchor_utils.py:127: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -291,7 +291,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 54.429 seconds)
+ **Total running time of the script:** ( 3 minutes 1.189 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 176e69ae0..c4ad86ed0 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -219,7 +219,7 @@ training. Other models require a full post training calibration.
.. code-block:: none
Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
-
0%| | 0.00/13.6M [00:00<?, ?B/s]
6%|6 | 880k/13.6M [00:00<00:01, 8.96MB/s]
18%|#8 | 2.48M/13.6M [00:00<00:00, 13.6MB/s]
40%|#### | 5.43M/13.6M [00:00<00:00, 21.5MB/s]
80%|#######9 | 10.8M/13.6M [00:00<00:00, 35.2MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 32.6MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
9%|9 | 1.23M/13.6M [00:00<00:01, 12.7MB/s]
27%|##7 | 3.73M/13.6M [00:00<00:00, 20.5MB/s]
65%|######4 | 8.75M/13.6M [00:00<00:00, 35.1MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 39.1MB/s]
@@ -399,7 +399,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.4646 90.2163 100.8701 90.0853 1.3153
+ 90.3201 90.2930 91.0299 90.1665 0.1299
@@ -448,7 +448,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 7.399 seconds)
+ **Total running time of the script:** ( 1 minutes 8.754 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 d9b28503e..cee2027eb 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
@@ -426,7 +426,7 @@ 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)
- 119.3751 119.3254 120.2596 118.6613 0.3022
+ 120.2101 120.1123 125.5592 119.5570 0.6454
@@ -463,7 +463,7 @@ Here we give an example of how to measure performance of TVM compiled models.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 51.697 seconds)
+ **Total running time of the script:** ( 1 minutes 51.255 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 353a22c7c..382350962 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -254,7 +254,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 29.035 seconds)
+ **Total running time of the script:** ( 1 minutes 11.584 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 4e7a9c8b9..d351aaeeb 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
@@ -157,7 +157,7 @@ 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 | 6010/132723 [00:00<00:02, 60095.90KB/s]
11%|# | 14441/132723 [00:00<00:01, 74337.36KB/s]
17%|#7 | 22880/132723 [00:00<00:01, 78925.54KB/s]
24%|##3 | 31323/132723 [00:00<00:01, 81095.09KB/s]
30%|### | 39842/132723 [00:00<00:01, 82569.00KB/s]
36%|###6 | 48277/132723 [00:00<00:01, 83171.91KB/s]
43%|####2 | 56678/132723 [00:00<00:00, 83443.31KB/s]
49%|####9 | 65163/132723 [00:00<00:00, 83888.20KB/s]
55%|#####5 | 73608/132723 [00:00<00:00, 84060.02KB/s]
62%|######1 | 82087/132723 [00:01<00:00, 84283.34KB/s]
68%|######8 | 90554/132723 [00:01<00:00, 84400.22KB/s]
75%|#######4 | 99024/132723 [00:01<00:00, 84484.57KB/s]
81%|######## | 107473/132723 [00:01<00:00, 84326.68KB/s]
87%|########7 | 115906/132723 [00:01<00:00, 84308.19KB/s]
94%|#########3| 124337/132723 [00:01<00:00, 84249.67KB/s]
100%|########
##| 132723/132723 [00:01<00:00, 82855.34KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
1%| | 940/132723 [00:00<00:14, 9328.09KB/s]
2%|1 | 2610/132723 [00:00<00:09, 13593.59KB/s]
4%|4 | 5746/132723 [00:00<00:05, 21630.93KB/s]
9%|8 | 11442/132723 [00:00<00:03, 35483.31KB/s]
15%|#4 | 19765/132723 [00:00<00:02, 52644.11KB/s]
21%|##1 | 28274/132723 [00:00<00:01, 63644.58KB/s]
28%|##7 | 36770/132723 [00:00<00:01, 70596.23KB/s]
34%|###4 | 45212/132723 [00:00<00:01, 74989.64KB/s]
41%|#### | 53778/132723 [00:00<00:01, 78320.42KB/s]
47%|####6 | 62329/132723 [00:01<00:00, 80536.89KB/s]
53%|#####3 | 70866/132723 [00:01<00:00, 82013.83KB/s]
60%|#####9 | 79371/132723 [00:01<00:00, 82934.93KB/s]
66%|######6 | 87915/132723 [00:01<00:00, 83691.48KB/s]
73%|#######2 | 96517/132723 [00:01<00:00, 84390.28KB/s]
79%|#######9 | 105074/132723 [00:01<00:00, 84742.93KB/s]
86%|########5 | 11
3604/132723 [00:01<00:00, 84908.60KB/s]
92%|#########2| 122134/132723 [00:01<00:00, 85023.80KB/s]
98%|#########8| 130706/132723 [00:01<00:00, 85224.95KB/s]
100%|##########| 132723/132723 [00:01<00:00, 72578.87KB/s]
@@ -240,7 +240,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 18.036 seconds)
+ **Total running time of the script:** ( 2 minutes 20.510 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 d82d36428..aa9af6728 100644
--- a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
@@ -5,22 +5,22 @@
Computation times
=================
-**10:31.098** total execution time for **how_to_deploy_models** files:
+**10:25.112** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 02:54.429 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:01.189 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 02:18.036 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 02:20.510 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 01:51.697 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 01:51.255 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:29.035 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:11.584 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:07.399 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:08.754 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:28.619 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:29.501 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:21.878 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:22.313 | 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 9183d07e4..858e25f32 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.zip4b5eceb2-88b7-4508-ac97-14ab3565b274 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip5813f13e-1029-4f7d-a687-2c9c2dbc1d30 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
@@ -577,7 +577,7 @@ Now, to actually convert the entire network, we have written `a pass in Relay <h
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- Check failed: (lower) is false: FloatImm lowering function for target llvm type 150 not found
+ Check failed: (lower) is false: Intrinsic lowering function for target llvm, intrinsic name tir.sqrt, type 150 not found
diff --git a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
index bc119cda8..ba3567d1e 100644
--- a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
@@ -5,14 +5,14 @@
Computation times
=================
-**00:39.774** total execution time for **how_to_extend_tvm** files:
+**00:40.139** total execution time for **how_to_extend_tvm** files:
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:36.589 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:36.585 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.257 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.639 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:00.922 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:00.908 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.006 | 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 04bcb918d..12cb3795a 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
@@ -215,10 +215,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 7271us [7271us] (46.43%; 46.43%)
- FoldScaleAxis: 8388us [6us] (53.57%; 53.57%)
- FoldConstant: 8382us [1669us] (53.53%; 99.93%)
- InferType: 6713us [6713us] (42.87%; 80.09%)
+ InferType: 6703us [6703us] (45.68%; 45.68%)
+ FoldScaleAxis: 7971us [6us] (54.32%; 54.32%)
+ FoldConstant: 7965us [1582us] (54.28%; 99.92%)
+ InferType: 6382us [6382us] (43.50%; 80.13%)
@@ -257,10 +257,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 6723us [6723us] (44.81%; 44.81%)
- FoldScaleAxis: 8281us [5us] (55.19%; 55.19%)
- FoldConstant: 8276us [1702us] (55.16%; 99.94%)
- InferType: 6574us [6574us] (43.81%; 79.43%)
+ InferType: 6341us [6341us] (44.56%; 44.56%)
+ FoldScaleAxis: 7890us [5us] (55.44%; 55.44%)
+ FoldConstant: 7885us [1612us] (55.41%; 99.94%)
+ InferType: 6273us [6273us] (44.08%; 79.56%)
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 dab80470b..6f6cb4cd2 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
@@ -327,7 +327,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 54.142719 ms
+ Convolution: 36.530970 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 7672c106a..0a2e8a782 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
@@ -658,7 +658,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 6.906179 ms
+ conv2d with tensor core: 13.162686 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 2c61b5545..ed85671e8 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -130,8 +130,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.019387
- Baseline: 3.341706
+ Numpy running time: 0.019243
+ Baseline: 3.344933
@@ -226,7 +226,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.307661
+ Opt1: 0.301809
@@ -329,7 +329,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.349376
+ Opt2: 0.344805
@@ -425,7 +425,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.118716
+ Opt3: 0.120931
@@ -550,7 +550,7 @@ flattening.
.. code-block:: none
- Opt4: 0.110635
+ Opt4: 0.111510
@@ -672,7 +672,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.111683
+ Opt5: 0.111780
@@ -797,7 +797,7 @@ Futhermore, we can also utilize multi-core processors to do the thread-level par
.. code-block:: none
- Opt6: 0.145276
+ Opt6: 0.145536
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 d437fbbb2..a2b41c29c 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
=================
-**00:34.598** total execution time for **how_to_optimize_operators** files:
+**00:34.594** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.358 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.242 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.233 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.318 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.007 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.033 | 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 a605a295f..31bedbfd2 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
@@ -5,18 +5,18 @@
Computation times
=================
-**05:22.741** total execution time for **how_to_tune_with_autoscheduler** files:
+**05:12.861** 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``) | 02:36.299 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 02:33.658 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:19.969 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:20.635 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 00:42.898 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 00:43.525 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:26.436 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:17.856 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:08.588 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:08.662 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:08.551 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:08.526 | 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 c18382f52..c2bc6a405 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
@@ -239,939 +239,659 @@ cooperative fetching, unrolling and operator fusion.
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 112;
- allocate(conv2d_nchw: Pointer(local float32), float32, [7]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [672]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [7], [], scope="local", align=16)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 64;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [8]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [648]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [576]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [8], [], scope="local", align=32)[0] = 0f32
conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[2] = 0f32
conv2d_nchw_1[3] = 0f32
conv2d_nchw_1[4] = 0f32
conv2d_nchw_1[5] = 0f32
conv2d_nchw_1[6] = 0f32
- for (rc.outer.outer: int32, 0, 16) {
- for (ry.outer.outer: int32, 0, 3) {
- let cse_var_3: int32 = (rc.outer.outer*1568)
- let cse_var_2: int32 = (ry.outer.outer*7)
- let cse_var_1: int32 = (ry.outer.outer*3)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32 {
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [672], [], scope="shared")[(threadIdx.x_1*8)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*8), 21), 3) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*8), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 1)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 1), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 1), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 1), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7 [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 2)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 2), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 2), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 2), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7 [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 3)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 1), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7)) + floormod((threa [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 4)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 4), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 4), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 4), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7 [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 5)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 5), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 5), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 5), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7 [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 6)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 2), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 2), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7)) + floormod((threa [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 7)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 7), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 7), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 7), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7 [...]
- }
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32 {
- pad_temp.shared_1[((threadIdx.x_1*8) + 256)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 256), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 256), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 256), 3)*7)) + cse_var_2) + floormod(block [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 257)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 257), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 257), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 257), 3)*7)) + cse_var_2) + floormod(block [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 258)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 2), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 2), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7)) + floormod((thr [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 259)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 256), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 256), 3) + 1), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 256), 3)*7)) + cse_var_2) + floo [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 260)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 260), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 260), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 260), 3)*7)) + cse_var_2) + floormod(block [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 261)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 3), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 3), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7)) + floormod((thr [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 262)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 256), 3) + 2), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 256), 3) + 2), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 256), 3)*7)) + cse_var_2) + floo [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 263)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 263), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 263), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 263), 3)*7)) + cse_var_2) + floormod(block [...]
- }
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32 {
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 512)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 512), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 512), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 512), 3)*7)) + cse_var_2) + floormod(blo [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 513)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 3), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 3), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7)) + floormod((t [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 514)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 514), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 514), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 514), 3)*7)) + cse_var_2) + floormod(blo [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 515)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 512), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 512), 3) + 1), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 512), 3)*7)) + cse_var_2) + fl [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 516)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 4), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 4), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7)) + floormod((t [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 517)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 517), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 517), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 517), 3)*7)) + cse_var_2) + floormod(blo [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 518)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 512), 3) + 2), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 512), 3) + 2), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 512), 3)*7)) + cse_var_2) + fl [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 519)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 5), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 5), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod(blockIdx.x, 7)) + floormod((t [...]
- }
- }
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope="shared")[threadIdx.x_2] = kernel[(((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 32)] = kernel[(((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[(((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 96)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 4608)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 4608)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 160)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 4608)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 9216)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 9216)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 9216)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 288)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 13824)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 13824)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 352)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 13824)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 18432)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 416)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 18432)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 18432)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 480)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 23040)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 23040)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 544)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 23040)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 27648)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 608)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 27648)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 27648)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 32256)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 32256)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 736)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 32256)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 800)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 864)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 41472)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 41472)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 928)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 41472)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 46080)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 992)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 46080)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 46080)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1056)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 50688)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 50688)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 50688)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 55296)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1184)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 55296)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 55296)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1248)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 59904)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 59904)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1312)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 59904)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 64512)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1376)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 64512)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 64512)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1440)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 69120)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 69120)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1504)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 69120)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1632)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 78336)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 78336)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1696)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 78336)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 82944)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1760)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 82944)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 82944)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1824)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 87552)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 87552)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1888)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 87552)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 92160)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1952)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 92160)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 92160)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 96768)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 96768)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2080)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 96768)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 101376)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2144)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 101376)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 101376)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2208)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 105984)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 105984)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2272)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 105984)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2336)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2400)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 115200)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 115200)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2464)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 115200)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 119808)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2528)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 119808)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 119808)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2592)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 124416)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 124416)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2656)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 124416)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 129024)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2720)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 129024)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 129024)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2784)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 133632)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 133632)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2848)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 133632)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 138240)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2912)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 138240)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 138240)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2976)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 142848)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 142848)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 3040)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 142848)]
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[0]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[9]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[12]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[18]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[72]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[81]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[80]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[90]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[99]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[89]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[98]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[108]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[117]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[107]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[116]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[125]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[126]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[135]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[144]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[134]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[143]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[153]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[162]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[152]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[161]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[171]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[180]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[170]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[179]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[188]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[189]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[198]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[207]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[197]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[206]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[216]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[219]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[222]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[225]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[228]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[217]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[220]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[223]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[226]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[229]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[215]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[218]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[221]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[224]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[227]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[230]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[231]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[234]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[237]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[240]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[243]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[246]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[249]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[232]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[235]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[238]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[241]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[244]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[247]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[250]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[233]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[236]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[239]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[242]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[245]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[248]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[251]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[252]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[255]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[258]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[261]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[264]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[267]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[270]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[253]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[256]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[259]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[262]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[265]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[268]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[271]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[254]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[257]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[260]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[263]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[266]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[269]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[272]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[273]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[276]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[279]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[282]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[285]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[288]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[291]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[274]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[277]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[280]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[283]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[286]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[289]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[292]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[275]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[278]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[281]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[284]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[287]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[290]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[293]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[294]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[297]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[300]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[303]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[306]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[309]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[312]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[295]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[298]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[301]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[304]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[307]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[310]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[313]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[296]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[299]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[302]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[305]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[308]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[311]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[314]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[315]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[318]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[321]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[324]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[327]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[330]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[333]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[316]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[319]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[322]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[325]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[328]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[331]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[334]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[317]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[320]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[323]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[326]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[329]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[332]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[335]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[336]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[339]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[342]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[345]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[348]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[351]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[354]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[337]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[340]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[343]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[346]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[349]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[352]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[355]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[338]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[341]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[344]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[347]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[350]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[353]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[356]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[357]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[360]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[363]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[366]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[369]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[372]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[375]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[358]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[361]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[364]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[367]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[370]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[373]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[376]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[359]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[362]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[365]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[368]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[371]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[374]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[377]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[378]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[381]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[384]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[387]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[390]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[393]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[396]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[379]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[382]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[385]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[388]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[391]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[394]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[397]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[380]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[383]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[386]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[389]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[392]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[395]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[398]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[399]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[402]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[405]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[408]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[411]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[414]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[417]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[400]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[403]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[406]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[409]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[412]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[415]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[418]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[401]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[404]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[407]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[410]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[413]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[416]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[419]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[420]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[423]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[426]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[429]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[432]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[435]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[438]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[421]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[424]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[427]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[430]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[433]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[436]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[439]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[422]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[425]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[428]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[431]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[434]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[437]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[440]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[441]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[444]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[447]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[450]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[453]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[456]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[459]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[442]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[445]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[448]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[451]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[454]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[457]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[460]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[443]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[446]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[449]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[452]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[455]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[458]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[461]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[462]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[465]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[468]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[471]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[474]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[477]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[480]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[463]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[466]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[469]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[472]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[475]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[478]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[481]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[464]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[467]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[470]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[473]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[476]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[479]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[482]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[483]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[486]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[489]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[492]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[495]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[498]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[501]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[484]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[487]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[490]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[493]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[496]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[499]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[502]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[485]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[488]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[491]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[494]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[497]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[500]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[503]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[504]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[507]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[510]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[513]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[516]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[519]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[522]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[505]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[508]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[511]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[514]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[517]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[520]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[523]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[506]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[509]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[512]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[515]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[518]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[521]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[524]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[525]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[528]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[531]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[534]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[537]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[540]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[543]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[526]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[529]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[532]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[535]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[538]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[541]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[544]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[527]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[530]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[533]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[536]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[539]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[542]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[545]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[546]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[549]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[552]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[555]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[558]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[561]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[564]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[547]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[550]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[553]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[556]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[559]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[562]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[565]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[548]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[551]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[554]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[557]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[560]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[563]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[566]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[567]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[570]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[573]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[576]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[579]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[582]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[585]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[568]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[571]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[574]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[577]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[580]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[583]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[586]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[569]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[572]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[575]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[578]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[581]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[584]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[587]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[588]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[591]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[594]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[597]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[600]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[603]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[606]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[589]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[592]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[595]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[598]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[601]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[604]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[607]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[590]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[593]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[596]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[599]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[602]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[605]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[608]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[609]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[612]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[615]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[618]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[621]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[624]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[627]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[610]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[613]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[616]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[619]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[622]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[625]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[628]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[611]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[614]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[617]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[620]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[623]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[626]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[629]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[630]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[633]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[636]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[639]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[642]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[645]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[648]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[631]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[634]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[637]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[640]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[643]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[646]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[649]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[632]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[635]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[638]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[641]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[644]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[647]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[650]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[651]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[654]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[657]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[660]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[663]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[666]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[669]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[652]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[655]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[658]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[661]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[664]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[667]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[670]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[653]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[656]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[659]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[662]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[665]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[668]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[671]*kernel.shared_1[((threadIdx.x*96) + 95)]))
+ conv2d_nchw_1[7] = 0f32
+ for (rc.outer.outer: int32, 0, 64) {
+ let cse_var_2: int32 = (rc.outer.outer*392)
+ let cse_var_1: int32 = (rc.outer.outer*72)
+ {
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [648], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else((((9 <= threadIdx.x_1) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[(((cse_var_2 + (floordiv(threadIdx.x_1, 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 49), 81)) && (floormod((threadIdx.x_1 + 49), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 49), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 49), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 8), 9)) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 98), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 98), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 147), 81)) && (floormod((threadIdx.x_1 + 66), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 3), 9))) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 147), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 147), 81), 9)*7)) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 196), 81)) && (floormod((threadIdx.x_1 + 34), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 196), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 196), 81), 9)*7)) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 245)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 245), 81)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 245), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 245), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 294), 81)) && (floormod((threadIdx.x_1 + 51), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 294), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 294), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 343)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 1), 9)) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 343), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 343), 81), 9)*7)) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 392), 81)) && (floormod((threadIdx.x_1 + 68), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 392), 81), 9)*7)) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 441)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 9) + 4), 9)) && (floormod((threadIdx.x_1 + 36), 81) < 72)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 441), 81)*49)) + (floormod((floordiv(threadIdx.x_1, 9) + 4), 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 490), 81)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 490), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 490), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 539)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 539), 81)) && (floormod((threadIdx.x_1 + 53), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 539), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 539), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 588)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 3), 9)) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 588), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 588), 81), 9)*7)) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ if @tir.likely((threadIdx.x_1 < 11), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 637)] = @tir.if_then_else((((floormod((threadIdx.x_1 + 70), 81) < 72) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 637), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 637), 81), 9)*7)) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
}
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1: Buffer(kernel.shared, float32, [576], [], scope="shared")[threadIdx.x_2] = kernel[(((blockIdx.x*36864) + cse_var_1) + threadIdx.x_2)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 49), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 49), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 98), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 98), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 147)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 147), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 1), 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 196)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 196), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 196), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 245)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 245), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 245), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 294)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 294), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 2), 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 343)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 343), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 343), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 392), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 392), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 441)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 441), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 3), 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 490)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 490), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 490), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ if @tir.likely((threadIdx.x_2 < 37), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 539)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 539), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 539), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ }
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[0]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[72]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[144]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[216]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[288]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[360]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[432]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[504]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[1]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[73]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[145]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[217]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[289]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[361]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[433]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[505]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[2]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[74]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[146]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[218]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[290]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[362]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[434]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[506]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[9]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[81]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[153]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[225]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[297]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[369]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[441]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[513]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[10]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[82]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[154]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[226]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[298]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[370]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[442]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[514]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[11]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[83]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[155]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[227]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[299]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[371]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[443]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[515]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[18]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[90]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[162]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[234]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[306]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[378]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[450]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[522]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[19]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[91]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[163]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[235]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[307]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[379]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[451]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[523]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[20]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[92]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[164]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[236]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[308]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[380]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[452]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[524]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[27]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[99]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[171]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[243]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[315]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[387]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[459]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[531]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[28]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[100]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[172]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[244]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[316]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[388]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[460]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[532]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[29]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[101]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[173]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[245]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[317]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[389]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[461]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[533]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[3]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[75]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[147]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[219]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[291]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[363]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[435]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[507]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[4]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[76]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[148]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[220]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[292]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[364]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[436]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[508]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[5]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[77]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[149]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[221]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[293]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[365]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[437]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[509]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[12]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[84]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[156]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[228]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[300]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[372]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[444]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[516]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[13]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[85]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[157]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[229]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[301]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[373]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[445]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[517]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[14]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[86]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[158]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[230]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[302]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[374]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[446]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[518]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[21]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[93]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[165]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[237]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[309]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[381]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[453]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[525]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[22]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[94]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[166]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[238]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[310]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[382]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[454]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[526]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[23]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[95]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[167]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[239]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[311]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[383]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[455]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[527]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[30]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[102]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[174]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[246]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[318]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[390]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[462]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[534]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[31]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[103]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[175]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[247]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[319]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[391]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[463]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[535]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[32]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[104]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[176]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[248]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[320]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[392]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[464]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[536]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[6]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[78]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[150]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[222]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[294]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[366]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[438]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[510]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[7]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[79]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[151]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[223]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[295]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[367]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[439]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[511]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[8]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[80]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[152]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[224]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[296]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[368]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[440]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[512]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[15]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[87]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[159]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[231]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[303]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[375]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[447]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[519]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[16]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[88]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[160]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[232]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[304]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[376]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[448]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[520]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[17]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[89]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[161]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[233]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[305]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[377]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[449]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[521]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[24]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[96]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[168]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[240]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[312]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[384]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[456]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[528]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[25]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[97]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[169]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[241]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[313]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[385]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[457]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[529]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[26]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[98]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[170]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[242]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[314]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[386]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[458]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[530]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[33]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[105]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[177]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[249]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[321]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[393]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[465]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[537]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[34]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[106]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[178]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[250]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[322]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[394]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[466]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[538]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[35]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[107]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[179]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[251]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[323]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[395]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[467]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[539]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[36]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[108]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[180]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[252]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[324]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[396]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[468]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[540]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[37]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[109]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[181]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[253]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[325]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[397]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[469]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[541]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[38]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[110]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[182]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[254]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[326]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[398]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[470]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[542]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[45]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[117]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[189]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[261]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[333]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[405]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[477]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[549]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[46]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[118]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[190]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[262]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[334]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[406]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[478]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[550]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[47]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[119]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[191]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[263]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[335]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[407]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[479]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[551]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[54]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[126]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[198]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[270]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[342]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[414]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[486]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[558]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[55]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[127]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[199]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[271]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[343]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[415]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[487]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[559]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[56]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[128]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[200]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[272]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[344]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[416]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[488]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[560]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[63]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[135]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[207]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[279]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[351]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[423]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[495]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[567]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[64]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[136]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[208]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[280]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[352]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[424]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[496]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[568]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[65]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[137]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[209]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[281]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[353]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[425]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[497]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[569]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[39]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[111]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[183]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[255]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[327]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[399]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[471]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[543]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[40]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[112]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[184]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[256]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[328]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[400]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[472]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[544]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[41]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[113]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[185]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[257]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[329]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[401]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[473]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[545]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[48]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[120]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[192]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[264]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[336]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[408]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[480]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[552]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[49]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[121]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[193]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[265]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[337]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[409]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[481]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[553]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[50]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[122]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[194]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[266]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[338]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[410]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[482]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[554]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[57]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[129]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[201]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[273]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[345]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[417]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[489]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[561]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[58]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[130]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[202]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[274]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[346]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[418]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[490]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[562]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[59]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[131]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[203]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[275]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[347]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[419]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[491]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[563]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[66]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[138]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[210]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[282]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[354]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[426]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[498]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[570]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[67]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[139]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[211]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[283]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[355]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[427]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[499]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[571]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[68]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[140]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[212]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[284]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[356]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[428]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[500]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[572]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[42]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[114]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[186]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[258]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[330]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[402]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[474]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[546]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[43]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[115]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[187]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[259]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[331]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[403]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[475]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[547]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[44]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[116]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[188]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[260]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[332]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[404]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[476]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[548]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[51]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[123]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[195]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[267]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[339]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[411]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[483]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[555]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[52]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[124]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[196]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[268]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[340]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[412]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[484]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[556]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[53]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[125]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[197]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[269]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[341]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[413]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[485]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[557]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[60]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[132]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[204]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[276]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[348]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[420]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[492]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[564]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[61]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[133]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[205]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[277]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[349]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[421]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[493]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[565]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[62]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[134]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[206]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[278]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[350]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[422]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[494]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[566]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[69]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[141]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[213]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[285]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[357]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[429]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[501]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[573]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[70]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[142]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[214]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[286]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[358]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[430]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[502]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[574]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[71]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[143]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[215]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[287]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[359]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[431]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[503]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[575]))
}
}
- for (i2.inner: int32, 0, 7) {
- compute[((((floordiv(blockIdx.x, 7)*1568) + (threadIdx.x*49)) + (i2.inner*7)) + floormod(blockIdx.x, 7))] = max((conv2d_nchw_1[i2.inner] + bias[((floordiv(blockIdx.x, 7)*32) + threadIdx.x)]), 0f32)
+ for (i1.inner: int32, 0, 8) {
+ compute[(((blockIdx.x*392) + (i1.inner*49)) + threadIdx.x)] = max((conv2d_nchw_1[i1.inner] + bias[((blockIdx.x*8) + i1.inner)]), 0f32)
}
}
}
@@ -1226,7 +946,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.404 ms
+ Execution time of this operator: 0.220 ms
@@ -1274,36 +994,36 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
- conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
+ conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=8)
conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
- conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=32)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=1)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
- conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=7)
+ 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=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=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=16)
+ conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
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=3)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
- compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=32)
+ 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=1)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
- compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=7)
- compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
+ compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+ compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
- compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=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)
@@ -1323,12 +1043,12 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=32)
+ 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=49)
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=8)
+ 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=32)
+ 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=49)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 1024)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
@@ -1348,10 +1068,10 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
#define int64_t long long
#define uint64_t unsigned long long
#endif
- extern "C" __global__ void __launch_bounds__(32) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[7];
- __shared__ float pad_temp_shared[672];
- __shared__ float kernel_shared[3072];
+ extern "C" __global__ void __launch_bounds__(49) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[8];
+ __shared__ float pad_temp_shared[648];
+ __shared__ float kernel_shared[576];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
@@ -1359,822 +1079,619 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw[4] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
- for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
- __syncthreads();
- pad_temp_shared[(((int)threadIdx.x) * 8)] = (((((1 <= ((((((int)threadIdx.x) * 8) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) * 8) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) * 2) % 3)) - 8 [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 1)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 1) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 1) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 2)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 2) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 2) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 3)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 1) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 4)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 4) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 4) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 5)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 5) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 5) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 5) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 6)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 2) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 2) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 7)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 7) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 7) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 7) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 256)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 4) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 256) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7 [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 257)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 5) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 5) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 257) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7 [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 258)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 2) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 2) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 259)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 8) + 256) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 8) + 256) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 256) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)b [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 260)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 8) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 8) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 260) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7 [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 261)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 3) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 3) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 262)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 8) + 256) / 3) + 2) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 8) + 256) / 3) + 2) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 256) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)b [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 263)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 11) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 11) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 263) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % [...]
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 512)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 8) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 8) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 512) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 513)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 3) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 3) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadI [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 514)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 10) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 10) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 514) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 515)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 8) + 512) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 8) + 512) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 512) / 3) * 7)) + (ry_outer_outer * 7)) + (((int [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 516)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 4) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 4) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadI [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 517)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 13) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 13) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 517) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 518)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 8) + 512) / 3) + 2) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 8) + 512) / 3) + 2) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 512) / 3) * 7)) + (ry_outer_outer * 7)) + (((int [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 519)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 5) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 5) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadI [...]
- }
- kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
- kernel_shared[(((int)threadIdx.x) + 32)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 64)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 96)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 4608)];
- kernel_shared[(((int)threadIdx.x) + 128)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 4608)];
- kernel_shared[(((int)threadIdx.x) + 160)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 4608)];
- kernel_shared[(((int)threadIdx.x) + 192)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 9216)];
- kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 9216)];
- kernel_shared[(((int)threadIdx.x) + 256)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 9216)];
- kernel_shared[(((int)threadIdx.x) + 288)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 13824)];
- kernel_shared[(((int)threadIdx.x) + 320)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 13824)];
- kernel_shared[(((int)threadIdx.x) + 352)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 13824)];
- kernel_shared[(((int)threadIdx.x) + 384)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 18432)];
- kernel_shared[(((int)threadIdx.x) + 416)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 18432)];
- kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 18432)];
- kernel_shared[(((int)threadIdx.x) + 480)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 23040)];
- kernel_shared[(((int)threadIdx.x) + 512)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 23040)];
- kernel_shared[(((int)threadIdx.x) + 544)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 23040)];
- kernel_shared[(((int)threadIdx.x) + 576)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 27648)];
- kernel_shared[(((int)threadIdx.x) + 608)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 27648)];
- kernel_shared[(((int)threadIdx.x) + 640)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 27648)];
- kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 32256)];
- kernel_shared[(((int)threadIdx.x) + 704)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 32256)];
- kernel_shared[(((int)threadIdx.x) + 736)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 32256)];
- kernel_shared[(((int)threadIdx.x) + 768)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((int)threadIdx.x) + 800)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 36864)];
- kernel_shared[(((int)threadIdx.x) + 832)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 36864)];
- kernel_shared[(((int)threadIdx.x) + 864)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 41472)];
- kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 41472)];
- kernel_shared[(((int)threadIdx.x) + 928)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 41472)];
- kernel_shared[(((int)threadIdx.x) + 960)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 46080)];
- kernel_shared[(((int)threadIdx.x) + 992)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 46080)];
- kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 46080)];
- kernel_shared[(((int)threadIdx.x) + 1056)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 50688)];
- kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 50688)];
- kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 50688)];
- kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 55296)];
- kernel_shared[(((int)threadIdx.x) + 1184)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 55296)];
- kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 55296)];
- kernel_shared[(((int)threadIdx.x) + 1248)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 59904)];
- kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 59904)];
- kernel_shared[(((int)threadIdx.x) + 1312)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 59904)];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 64512)];
- kernel_shared[(((int)threadIdx.x) + 1376)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 64512)];
- kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 64512)];
- kernel_shared[(((int)threadIdx.x) + 1440)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 69120)];
- kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 69120)];
- kernel_shared[(((int)threadIdx.x) + 1504)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 69120)];
- kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 73728)];
- kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 73728)];
- kernel_shared[(((int)threadIdx.x) + 1632)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 78336)];
- kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 78336)];
- kernel_shared[(((int)threadIdx.x) + 1696)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 78336)];
- kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 82944)];
- kernel_shared[(((int)threadIdx.x) + 1760)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 82944)];
- kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 82944)];
- kernel_shared[(((int)threadIdx.x) + 1824)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 87552)];
- kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 87552)];
- kernel_shared[(((int)threadIdx.x) + 1888)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 87552)];
- kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 92160)];
- kernel_shared[(((int)threadIdx.x) + 1952)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 92160)];
- kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 92160)];
- kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 96768)];
- kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 96768)];
- kernel_shared[(((int)threadIdx.x) + 2080)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 96768)];
- kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 101376)];
- kernel_shared[(((int)threadIdx.x) + 2144)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 101376)];
- kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 101376)];
- kernel_shared[(((int)threadIdx.x) + 2208)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 105984)];
- kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 105984)];
- kernel_shared[(((int)threadIdx.x) + 2272)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 105984)];
- kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((int)threadIdx.x) + 2336)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 110592)];
- kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 110592)];
- kernel_shared[(((int)threadIdx.x) + 2400)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 115200)];
- kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 115200)];
- kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 115200)];
- kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 119808)];
- kernel_shared[(((int)threadIdx.x) + 2528)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 119808)];
- kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 119808)];
- kernel_shared[(((int)threadIdx.x) + 2592)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 124416)];
- kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 124416)];
- kernel_shared[(((int)threadIdx.x) + 2656)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 124416)];
- kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 129024)];
- kernel_shared[(((int)threadIdx.x) + 2720)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 129024)];
- kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 129024)];
- kernel_shared[(((int)threadIdx.x) + 2784)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 133632)];
- kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 133632)];
- kernel_shared[(((int)threadIdx.x) + 2848)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 133632)];
- kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 138240)];
- kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 138240)];
- kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 138240)];
- kernel_shared[(((int)threadIdx.x) + 2976)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 142848)];
- kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 142848)];
- kernel_shared[(((int)threadIdx.x) + 3040)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 142848)];
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[9] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[12] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[18] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[72] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[81] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[80] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[90] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[99] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[89] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[98] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[108] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[117] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[107] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[116] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[125] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[126] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[135] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[144] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[134] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[143] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[153] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[162] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[152] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[161] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[171] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[180] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[170] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[179] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[188] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[189] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[198] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[207] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[197] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[206] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[216] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[219] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[222] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[225] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[228] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[217] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[220] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[223] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[226] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[229] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[215] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[218] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[221] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[224] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[227] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[230] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[231] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[234] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[237] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[240] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[243] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[246] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[249] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[232] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[235] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[238] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[241] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[244] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[247] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[250] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[233] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[236] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[239] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[242] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[245] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[248] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[251] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[252] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[255] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[258] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[261] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[264] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[267] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[270] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[253] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[256] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[259] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[262] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[265] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[268] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[271] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[254] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[257] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[260] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[263] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[266] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[269] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[272] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[273] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[276] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[279] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[282] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[285] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[288] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[291] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[274] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[277] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[280] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[283] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[286] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[289] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[292] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[275] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[278] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[281] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[284] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[287] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[290] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[293] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[294] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[297] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[300] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[303] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[306] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[309] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[312] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[295] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[298] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[301] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[304] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[307] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[310] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[313] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[296] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[299] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[302] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[305] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[308] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[311] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[314] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[315] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[318] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[321] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[324] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[327] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[330] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[333] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[316] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[319] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[322] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[325] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[328] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[331] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[334] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[317] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[320] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[323] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[326] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[329] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[332] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[335] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[336] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[339] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[342] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[345] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[348] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[351] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[354] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[337] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[340] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[343] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[346] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[349] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[352] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[355] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[338] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[341] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[344] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[347] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[350] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[353] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[356] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[357] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[360] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[363] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[366] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[369] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[372] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[375] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[358] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[361] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[364] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[367] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[370] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[373] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[376] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[359] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[362] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[365] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[368] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[371] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[374] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[377] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[378] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[381] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[384] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[387] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[390] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[393] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[396] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[379] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[382] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[385] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[388] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[391] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[394] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[397] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[380] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[383] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[386] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[389] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[392] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[395] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[398] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[399] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[402] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[405] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[408] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[411] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[414] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[417] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[400] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[403] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[406] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[409] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[412] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[415] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[418] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[401] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[404] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[407] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[410] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[413] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[416] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[419] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[420] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[423] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[426] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[429] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[432] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[435] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[438] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[421] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[424] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[427] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[430] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[433] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[436] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[439] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[422] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[425] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[428] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[431] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[434] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[437] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[440] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[441] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[444] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[447] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[450] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[453] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[456] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[459] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[442] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[445] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[448] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[451] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[454] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[457] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[460] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[443] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[446] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[449] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[452] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[455] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[458] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[461] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[462] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[465] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[468] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[471] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[474] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[477] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[480] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[463] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[466] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[469] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[472] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[475] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[478] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[481] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[464] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[467] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[470] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[473] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[476] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[479] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[482] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[483] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[486] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[489] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[492] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[495] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[498] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[501] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[484] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[487] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[490] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[493] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[496] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[499] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[502] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[485] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[488] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[491] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[494] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[497] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[500] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[503] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[504] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[507] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[510] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[513] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[516] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[519] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[522] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[505] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[508] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[511] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[514] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[517] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[520] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[523] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[506] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[509] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[512] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[515] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[518] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[521] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[524] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[525] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[528] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[531] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[534] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[537] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[540] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[543] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[526] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[529] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[532] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[535] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[538] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[541] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[544] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[527] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[530] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[533] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[536] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[539] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[542] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[545] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[546] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[549] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[552] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[555] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[558] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[561] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[564] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[547] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[550] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[553] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[556] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[559] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[562] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[565] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[548] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[551] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[554] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[557] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[560] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[563] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[566] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[567] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[570] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[573] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[576] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[579] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[582] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[585] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[568] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[571] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[574] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[577] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[580] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[583] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[586] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[569] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[572] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[575] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[578] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[581] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[584] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[587] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[588] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[591] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[594] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[597] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[600] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[603] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[606] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[589] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[592] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[595] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[598] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[601] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[604] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[607] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[590] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[593] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[596] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[599] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[602] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[605] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[608] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[609] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[612] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[615] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[618] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[621] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[624] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[627] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[610] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[613] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[616] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[619] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[622] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[625] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[628] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[611] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[614] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[617] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[620] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[623] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[626] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[629] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[630] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[633] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[636] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[639] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[642] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[645] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[648] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[631] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[634] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[637] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[640] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[643] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[646] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[649] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[632] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[635] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[638] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[641] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[644] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[647] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[650] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[651] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[654] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[657] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[660] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[663] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[666] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[669] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[652] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[655] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[658] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[661] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[664] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[667] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[670] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[653] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[656] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[659] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[662] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[665] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[668] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[671] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
+ conv2d_nchw[7] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[((int)threadIdx.x)] = ((((9 <= ((int)threadIdx.x)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 49)] = (((((9 <= ((((int)threadIdx.x) + 49) % 81)) && (((((int)threadIdx.x) + 49) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 49) / 81) * 49)) + ((((((int)threadIdx.x) + 49) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 98)] = (((1 <= ((((int)threadIdx.x) + 8) % 9)) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((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) + 147)] = (((((9 <= ((((int)threadIdx.x) + 66) % 81)) && (((((int)threadIdx.x) + 66) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 147) / 81) * 49)) + ((((((int)threadIdx.x) + 66) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 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 * 392) + (((((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) + 245)] = ((((9 <= ((((int)threadIdx.x) + 2) % 81)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 245) / 81) * 49)) + ((((((int)threadIdx.x) + 2) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 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 * 392) + (((((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) + 343)] = (((1 <= ((((int)threadIdx.x) + 1) % 9)) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 343) / 81) * 49)) + ((((((int)threadIdx.x) + 19) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 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 * 392) + (((((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) + 441)] = (((((1 <= (((((int)threadIdx.x) / 9) + 4) % 9)) && (((((int)threadIdx.x) + 36) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 441) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 4) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 490)] = ((((9 <= ((((int)threadIdx.x) + 4) % 81)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((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) + 539)] = (((((9 <= ((((int)threadIdx.x) + 53) % 81)) && (((((int)threadIdx.x) + 53) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 539) / 81) * 49)) + ((((((int)threadIdx.x) + 53) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 588)] = (((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 588) / 81) * 49)) + ((((((int)threadIdx.x) + 21) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 11) {
+ pad_temp_shared[(((int)threadIdx.x) + 637)] = ((((((int)threadIdx.x) < 2) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 637) / 81) * 49)) + ((((((int)threadIdx.x) + 70) % 81) / 9) * 7)) + (((int)threadIdx.x) + 7)) - 8)] : 0.000000e+00f);
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 36864) + (rc_outer_outer * 72)) + ((int)threadIdx.x))];
+ kernel_shared[(((int)threadIdx.x) + 49)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 49) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 49) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 98)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 98) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 26) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 147)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 147) / 72) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) / 3) + 1) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 196) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 52) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 245)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 245) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 29) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 294)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 294) / 72) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) / 3) + 2) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 343)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 343) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 55) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 392) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 32) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 441)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 441) / 72) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) / 3) + 3) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 490)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 490) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 58) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ if (((int)threadIdx.x) < 37) {
+ kernel_shared[(((int)threadIdx.x) + 539)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 539) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 35) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
}
+ __syncthreads();
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[0]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[72]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[144]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[216]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[288]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[360]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[432]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[504]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[1]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[73]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[145]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[217]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[289]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[361]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[433]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[505]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[2]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[74]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[146]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[218]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[290]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[362]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[434]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[506]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[9]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[81]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[153]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[225]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[297]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[369]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[441]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[513]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[10]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[82]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[154]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[226]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[298]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[370]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[442]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[514]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[11]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[83]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[155]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[227]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[299]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[371]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[443]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[515]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[18]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[90]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[162]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[234]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[306]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[378]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[450]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[522]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[19]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[91]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[163]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[235]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[307]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[379]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[451]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[523]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[20]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[92]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[164]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[236]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[308]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[380]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[452]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[524]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[27]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[99]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[171]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[243]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[315]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[387]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[459]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[531]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[28]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[100]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[172]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[244]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[316]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[388]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[460]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[532]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[29]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[101]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[173]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[245]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[317]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[389]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[461]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[533]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[3]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[75]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[147]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[219]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[291]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[363]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[435]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[507]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[4]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[76]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[148]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[220]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[292]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[364]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[436]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[508]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[5]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[77]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[149]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[221]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[293]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[365]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[437]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[509]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[12]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[84]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[156]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[228]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[300]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[372]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[444]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[516]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[13]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[85]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[157]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[229]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[301]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[373]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[445]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[517]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[14]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[86]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[158]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[230]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[302]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[374]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[446]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[518]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[21]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[93]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[165]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[237]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[309]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[381]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[453]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[525]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[22]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[94]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[166]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[238]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[310]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[382]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[454]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[526]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[23]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[95]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[167]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[239]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[311]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[383]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[455]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[527]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[30]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[102]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[174]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[246]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[318]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[390]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[462]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[534]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[31]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[103]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[175]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[247]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[319]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[391]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[463]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[535]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[32]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[104]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[176]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[248]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[320]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[392]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[464]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[536]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[6]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[78]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[150]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[222]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[294]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[366]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[438]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[510]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[7]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[79]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[151]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[223]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[295]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[367]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[439]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[511]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[8]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[80]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[152]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[224]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[296]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[368]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[440]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[512]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[15]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[87]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[159]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[231]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[303]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[375]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[447]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[519]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[16]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[88]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[160]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[232]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[304]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[376]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[448]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[520]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[17]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[89]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[161]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[233]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[305]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[377]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[449]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[521]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[24]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[96]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[168]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[240]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[312]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[384]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[456]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[528]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[25]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[97]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[169]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[241]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[313]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[385]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[457]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[529]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[26]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[98]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[170]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[242]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[314]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[386]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[458]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[530]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[33]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[105]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[177]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[249]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[321]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[393]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[465]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[537]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[34]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[106]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[178]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[250]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[322]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[394]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[466]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[538]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[35]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[107]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[179]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[251]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[323]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[395]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[467]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[539]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[36]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[108]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[180]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[252]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[324]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[396]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[468]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[540]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[37]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[109]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[181]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[253]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[325]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[397]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[469]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[541]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[38]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[110]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[182]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[254]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[326]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[398]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[470]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[542]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[45]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[117]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[189]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[261]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[333]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[405]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[477]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[549]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[46]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[118]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[190]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[262]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[334]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[406]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[478]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[550]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[47]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[119]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[191]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[263]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[335]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[407]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[479]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[551]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[54]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[126]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[198]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[270]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[342]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[414]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[486]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[558]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[55]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[127]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[199]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[271]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[343]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[415]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[487]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[559]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[56]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[128]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[200]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[272]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[344]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[416]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[488]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[560]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[63]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[135]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[207]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[279]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[351]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[423]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[495]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[567]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[64]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[136]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[208]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[280]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[352]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[424]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[496]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[568]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[65]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[137]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[209]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[281]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[353]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[425]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[497]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[569]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[39]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[111]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[183]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[255]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[327]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[399]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[471]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[543]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[40]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[112]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[184]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[256]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[328]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[400]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[472]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[544]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[41]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[113]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[185]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[257]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[329]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[401]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[473]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[545]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[48]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[120]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[192]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[264]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[336]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[408]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[480]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[552]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[49]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[121]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[193]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[265]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[337]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[409]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[481]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[553]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[50]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[122]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[194]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[266]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[338]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[410]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[482]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[554]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[57]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[129]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[201]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[273]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[345]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[417]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[489]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[561]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[58]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[130]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[202]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[274]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[346]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[418]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[490]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[562]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[59]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[131]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[203]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[275]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[347]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[419]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[491]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[563]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[66]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[138]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[210]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[282]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[354]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[426]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[498]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[570]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[67]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[139]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[211]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[283]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[355]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[427]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[499]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[571]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[68]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[140]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[212]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[284]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[356]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[428]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[500]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[572]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[42]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[114]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[186]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[258]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[330]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[402]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[474]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[546]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[43]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[115]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[187]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[259]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[331]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[403]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[475]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[547]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[44]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[116]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[188]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[260]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[332]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[404]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[476]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[548]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[51]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[123]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[195]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[267]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[339]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[411]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[483]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[555]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[52]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[124]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[196]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[268]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[340]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[412]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[484]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[556]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[53]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[125]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[197]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[269]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[341]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[413]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[485]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[557]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[60]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[132]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[204]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[276]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[348]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[420]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[492]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[564]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[61]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[133]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[205]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[277]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[349]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[421]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[493]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[565]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[62]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[134]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[206]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[278]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[350]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[422]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[494]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[566]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[69]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[141]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[213]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[285]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[357]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[429]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[501]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[573]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[70]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[142]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[214]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[286]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[358]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[430]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[502]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[574]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[71]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[143]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[215]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[287]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[359]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[431]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[503]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[575]));
}
- for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
- compute[(((((((int)blockIdx.x) / 7) * 1568) + (((int)threadIdx.x) * 49)) + (i2_inner * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i2_inner] + bias[(((((int)blockIdx.x) / 7) * 32) + ((int)threadIdx.x))]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 8; ++i1_inner) {
+ compute[(((((int)blockIdx.x) * 392) + (i1_inner * 49)) + ((int)threadIdx.x))] = max((conv2d_nchw[i1_inner] + bias[((((int)blockIdx.x) * 8) + i1_inner)]), 0.000000e+00f);
}
}
@@ -2236,7 +1753,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 36.299 seconds)
+ **Total running time of the script:** ( 2 minutes 33.658 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 324e03dd2..cc3f639f7 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
@@ -646,7 +646,7 @@ 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)
- 10.1770 10.1816 10.2300 10.1194 0.0453
+ 9.8357 9.8624 9.8666 9.7782 0.0407
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 c909151cc..e0e77acf0 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
@@ -665,7 +665,7 @@ 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)
- 756.4694 756.3174 758.9303 754.1606 1.9502
+ 764.1533 763.5268 766.3735 762.5596 1.6188
@@ -693,7 +693,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 19.969 seconds)
+ **Total running time of the script:** ( 1 minutes 20.635 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 f9a902980..9bcf76e69 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
@@ -396,27 +396,102 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_9: placeholder_15: Buffer(placeholder_14, float32, [128, 512], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_5: placeholder_17: Buffer(placeholder_10, float32, [128, 256], []), placeholder_7: placeholder_18: Buffer(placeholder_12, int32, [4916], []), placeholder_8: placeholder_19: Buffer(placeholder_13, int32, [33], [])} {
- for (i0.outer: int32, 0, 64) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [64]), storage_scope = global;
- for (i1.outer: int32, 0, 16) {
- for (i.outer.inner: int32, 0, 2) {
- for (nb_j.inner: int32, 0, 2) {
- for (j.init: int32, 0, 16) {
- compute_5: Buffer(compute_4, float32, [64], [])[(((i.outer.inner*32) + (nb_j.inner*16)) + j.init)] = 0f32
+ preflattened_buffer_map = {placeholder_5: placeholder_15: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_7: placeholder_18: Buffer(placeholder_12, int32, [4916], []), placeholder_8: placeholder_19: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], [])} {
+ for (i0.outer: int32, 0, 16) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [128]), storage_scope = global;
+ for (i1.outer: int32, 0, 32) {
+ for (i.inner.init: int32, 0, 8) {
+ let cse_var_1: int32 = (i.inner.init*16)
+ {
+ compute_5: Buffer(compute_4, float32, [128], [])[cse_var_1] = 0f32
+ compute_5[(cse_var_1 + 1)] = 0f32
+ compute_5[(cse_var_1 + 2)] = 0f32
+ compute_5[(cse_var_1 + 3)] = 0f32
+ compute_5[(cse_var_1 + 4)] = 0f32
+ compute_5[(cse_var_1 + 5)] = 0f32
+ compute_5[(cse_var_1 + 6)] = 0f32
+ compute_5[(cse_var_1 + 7)] = 0f32
+ compute_5[(cse_var_1 + 8)] = 0f32
+ compute_5[(cse_var_1 + 9)] = 0f32
+ compute_5[(cse_var_1 + 10)] = 0f32
+ compute_5[(cse_var_1 + 11)] = 0f32
+ compute_5[(cse_var_1 + 12)] = 0f32
+ compute_5[(cse_var_1 + 13)] = 0f32
+ compute_5[(cse_var_1 + 14)] = 0f32
+ compute_5[(cse_var_1 + 15)] = 0f32
+ }
+ }
+ for (elem_idx: int32, 0, (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])) {
+ for (i.inner: int32, 0, 8) {
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_2: int32 = (i.inner*16)
+ compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[i1.outer]*16) + (elem_idx*16))]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_3: int32 = ((i.inner*16) + 1)
+ compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_4: int32 = ((i.inner*16) + 2)
+ compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_5: int32 = ((i.inner*16) + 3)
+ compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_6: int32 = ((i.inner*16) + 4)
+ compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_7: int32 = ((i.inner*16) + 5)
+ compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_8: int32 = ((i.inner*16) + 6)
+ compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_9: int32 = ((i.inner*16) + 7)
+ compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_10: int32 = ((i.inner*16) + 8)
+ compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_11: int32 = ((i.inner*16) + 9)
+ compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_12: int32 = ((i.inner*16) + 10)
+ compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_13: int32 = ((i.inner*16) + 11)
+ compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_14: int32 = ((i.inner*16) + 12)
+ compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_15: int32 = ((i.inner*16) + 13)
+ compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_16: int32 = ((i.inner*16) + 14)
+ compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
}
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((i1.outer*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((i1.outer*2) + nb_j.inner)
- let cse_var_2: int32 = (((i.outer.inner*32) + (nb_j.inner*16)) + j)
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((i0.outer*512) + (i.outer.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
- }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_17: int32 = ((i.inner*16) + 15)
+ compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
}
}
}
- for (i0.inner: int32, 0, 2) {
- let cse_var_4: int32 = (((i0.outer*1024) + (i0.inner*512)) + (i1.outer*32))
- compute[ramp(cse_var_4, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 8) {
+ let cse_var_18: int32 = (((i0.outer*4096) + (i0.inner*512)) + (i1.outer*16))
+ compute[ramp(cse_var_18, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_18, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -472,7 +547,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 1.902 ms
+ Execution time of this operator: 1.829 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 ffcaeaaef..b0da1db82 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
=================
-**00:43.177** total execution time for **how_to_tune_with_autotvm** files:
+**00:43.701** 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``) | 00:43.145 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:43.664 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.019 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.022 | 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_cuda.py` (``tune_relay_cuda.py``) | 00:00.005 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``) | 00:00.004 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``) | 00:00.004 | 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 0aba27573..f2f0c7228 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
@@ -879,8 +879,8 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2885496
- No: 6 GFLOPS: 109.11/109.11 result: MeasureResult(costs=(0.002121732916666667,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6635711193084717, timestamp=1656094762.5332358) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
- No: 7 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 6 GFLOPS: 110.60/110.60 result: MeasureResult(costs=(0.0020930916875,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8403575420379639, timestamp=1656095896.9995742) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
+ No: 7 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1003,7 +1003,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6225319
- No: 8 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 8 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1126,7 +1126,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,943546
- No: 9 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 9 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1249,7 +1249,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2868708
- No: 10 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 10 GFLOPS: 0.00/110.60 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
@@ -1267,7 +1267,7 @@ for this template
TimeoutError
[('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4691833
- No: 11 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 11 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1390,7 +1390,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1042124
- No: 12 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 12 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1513,7 +1513,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10013405
- No: 13 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 13 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1636,7 +1636,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6732082
- No: 14 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 14 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1759,7 +1759,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7536735
- No: 15 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 15 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1882,7 +1882,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,482121
- No: 16 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 16 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2005,7 +2005,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2824525
- No: 17 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 17 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2128,7 +2128,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4559286
- No: 18 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 18 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2251,7 +2251,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9677544
- No: 19 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+ No: 19 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 738, in __call__
yield remote, remote.load_module(os.path.split(build_result.filename)[1])
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 702, in run_through_rpc
@@ -2339,7 +2339,7 @@ for this template
15: _PyEval_EvalFrameDefault
14: 0x0000000000537c30
13: _PyObject_FastCallKeywords
- 12: 0x00007f7e217a3fa2
+ 12: 0x00007fb124dccfa2
11: _ctypes_callproc
10: ffi_call
9: ffi_call_unix64
@@ -2404,7 +2404,7 @@ for this template
21: _PyFunction_FastCallKeywords
20: _PyEval_EvalFrameDefault
19: _PyFunction_FastCall [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6390073
- No: 20 GFLOPS: 141.78/141.78 result: MeasureResult(costs=(0.0016327872741935482,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.1234588623046875, timestamp=1656094788.7532756) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
+ No: 20 GFLOPS: 144.84/144.84 result: MeasureResult(costs=(0.0015983578700000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4103615283966064, timestamp=1656095923.5313025) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
@@ -2461,7 +2461,7 @@ and measure running time.
Best config:
[('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
Finish loading 20 records
- Time cost of this operator: 0.002009
+ Time cost of this operator: 0.001963
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 4a8cca93e..e79bf831c 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
@@ -328,10 +328,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.7 98.727 (1, 2, 10, 10, 3) 2 1
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.106 0.981 (1, 6, 10, 10) 1 1
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.927 0.293 (1, 1, 10, 10, 3) 1 1
- Total_time - 316.733 - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 315.4 98.746 (1, 2, 10, 10, 3) 2 1
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.085 0.966 (1, 6, 10, 10) 1 1
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.921 0.288 (1, 1, 10, 10, 3) 1 1
+ Total_time - 319.406 - - - -
@@ -397,10 +397,10 @@ Timing the tuned program
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 192.2 98.556 (1, 1, 10, 10, 6) 2 1
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.982 1.017 (1, 6, 10, 10) 1 1
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.834 0.428 (1, 3, 10, 10, 1) 1 1
- Total_time - 195.017 - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 89.5 97.089 (1, 6, 10, 10, 1) 2 1
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.761 1.911 (1, 6, 10, 10) 1 1
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.922 1.0 (1, 1, 10, 10, 3) 1 1
+ Total_time - 92.183 - - - -
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 828fdb2fb..aa9e70b7e 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
@@ -225,7 +225,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
.. code-block:: none
- '/tmp/tmpcbeckp09/images/random'
+ '/tmp/tmp2ygzw1mt/images/random'
@@ -325,8 +325,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmpcbeckp09/images/target contains 8144 images
- /tmp/tmpcbeckp09/images/random contains 5000 images
+ /tmp/tmp2ygzw1mt/images/target contains 8144 images
+ /tmp/tmp2ygzw1mt/images/random contains 5000 images
@@ -501,13 +501,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 55s - loss: 0.2223 - accuracy: 0.9234 - val_loss: 0.1299 - val_accuracy: 0.9596
+ 328/328 - 57s - loss: 0.2227 - accuracy: 0.9251 - val_loss: 0.1449 - val_accuracy: 0.9577
Epoch 2/3
- 328/328 - 52s - loss: 0.1018 - accuracy: 0.9615 - val_loss: 0.1280 - val_accuracy: 0.9615
+ 328/328 - 53s - loss: 0.0938 - accuracy: 0.9636 - val_loss: 0.1395 - val_accuracy: 0.9588
Epoch 3/3
- 328/328 - 52s - loss: 0.0694 - accuracy: 0.9742 - val_loss: 0.1528 - val_accuracy: 0.9452
+ 328/328 - 53s - loss: 0.0669 - accuracy: 0.9755 - val_loss: 0.1180 - val_accuracy: 0.9641
- <keras.callbacks.History object at 0x7f675d8a59d0>
+ <keras.callbacks.History object at 0x7f87ac6b3c90>
@@ -864,7 +864,7 @@ Arduino tutorial for how to do that `on GitHub <https://github.com/guberti/tvm-a
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 7 minutes 59.583 seconds)
+ **Total running time of the script:** ( 7 minutes 49.189 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 da2bb2026..6f71126ec 100644
--- a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
@@ -5,14 +5,14 @@
Computation times
=================
-**08:45.837** total execution time for **how_to_work_with_microtvm** files:
+**08:36.560** total execution time for **how_to_work_with_microtvm** files:
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 07:59.583 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 07:49.189 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:42.799 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:43.769 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.455 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.601 | 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 f49b7d3c3..5add47d33 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
@@ -5,12 +5,12 @@
Computation times
=================
-**00:09.539** total execution time for **how_to_work_with_relay** files:
+**00:13.029** total execution time for **how_to_work_with_relay** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:08.075 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:10.510 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.459 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:02.513 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``) | 00:00.006 | 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 27f206fb3..1ae887c12 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
@@ -259,7 +259,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
.. code-block:: none
- <function my_cuda_math_rule at 0x7f66d2f2fb90>
+ <function my_cuda_math_rule at 0x7f8724cb1170>
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 242df368d..175fc36c6 100644
--- a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
@@ -5,22 +5,22 @@
Computation times
=================
-**00:04.018** total execution time for **how_to_work_with_schedules** files:
+**00:04.304** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:01.886 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:01.902 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:00.920 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.177 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.524 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.528 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.517 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.512 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.098 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.105 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.033 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.034 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.028 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.031 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.012 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.013 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
index c778109a1..3a365b0cc 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -346,7 +346,7 @@ The importing needs to happen before the tensorized GEMV being executed.
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpt4zzdkfh/input0.cc'\nsource_filename = \"/tmp/tmpt4zzdkfh/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpdpnmy4_h/input0.cc'\nsource_filename = \"/tmp/tmpdpnmy4_h/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index 5f4ee75e6..5428c064f 100644
--- a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:20.783** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:20.779** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:20.777 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:20.773 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``) | 00:00.006 | 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 bcad68249..014269c9b 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -291,7 +291,7 @@ The compilation steps are:
DeprecationWarning,
/workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the new recommended usage.
relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
- resnet18_v1 inference graph built in 22.49s!
+ resnet18_v1 inference graph built in 22.56s!
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 ed47328b3..7bb710b6b 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -335,7 +335,7 @@ The compilation steps are:
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/relay/build_module.py:411: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 15.73s!
+ yolov3-tiny inference graph built in 15.63s!
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 39ecf8611..145b67976 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
=================
-**01:30.302** total execution time for **topic_vta_tutorials_frontend** files:
+**01:30.977** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:47.768 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:47.688 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:42.534 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:43.288 | 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 3a047f86b..0fa510fdf 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
=================
-**00:03.268** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.190** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.882 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.802 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.386 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.388 | 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 5e7769a32..49624205f 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
=================
-**00:00.700** total execution time for **topic_vta_tutorials** files:
+**00:00.738** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.371 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.407 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.329 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.331 | 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 9edd66a6b..4daf279d6 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -327,7 +327,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 93.341 ms
+ Execution time of this operator: 94.166 ms
diff --git a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
index d4141aa3a..833814fcc 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -449,16 +449,16 @@ reduce variance, we take 5 measurements and average them.
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 10.34/10.34 result: MeasureResult(costs=(0.025971265799999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5495269298553467, timestamp=1656093636.995069) [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
- No: 2 GFLOPS: 2.65/10.34 result: MeasureResult(costs=(0.10123502820000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.765075922012329, timestamp=1656093638.7738595) [('tile_y', [-1, 4]), ('tile_x', [-1, 8])],None,32
- No: 3 GFLOPS: 11.79/11.79 result: MeasureResult(costs=(0.022772455400000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.55118727684021, timestamp=1656093639.8189595) [('tile_y', [-1, 64]), ('tile_x', [-1, 32])],None,56
- No: 4 GFLOPS: 1.56/11.79 result: MeasureResult(costs=(0.172348963,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.8733017444610596, timestamp=1656093643.256166) [('tile_y', [-1, 1]), ('tile_x', [-1, 4])],None,20
- No: 5 GFLOPS: 3.60/11.79 result: MeasureResult(costs=(0.0744695276,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3304519653320312, timestamp=1656093644.7122812) [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
- No: 6 GFLOPS: 1.78/11.79 result: MeasureResult(costs=(0.15104362659999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.578220844268799, timestamp=1656093647.3361433) [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
- No: 7 GFLOPS: 0.82/11.79 result: MeasureResult(costs=(0.3274878444,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.3515944480896, timestamp=1656093653.259746) [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
- No: 8 GFLOPS: 10.04/11.79 result: MeasureResult(costs=(0.0267448458,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5762312412261963, timestamp=1656093653.84942) [('tile_y', [-1, 4]), ('tile_x', [-1, 64])],None,62
- No: 9 GFLOPS: 1.68/11.79 result: MeasureResult(costs=(0.1596127274,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.651829481124878, timestamp=1656093656.6200204) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
- No: 10 GFLOPS: 2.26/11.79 result: MeasureResult(costs=(0.11855880939999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.004706621170044, timestamp=1656093658.684421) [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
+ No: 1 GFLOPS: 9.15/9.15 result: MeasureResult(costs=(0.029337883,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6015806198120117, timestamp=1656094754.0304887) [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
+ No: 2 GFLOPS: 2.42/9.15 result: MeasureResult(costs=(0.11107918779999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9320051670074463, timestamp=1656094755.977225) [('tile_y', [-1, 4]), ('tile_x', [-1, 8])],None,32
+ No: 3 GFLOPS: 11.79/11.79 result: MeasureResult(costs=(0.0227689394,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5547053813934326, timestamp=1656094757.0288591) [('tile_y', [-1, 64]), ('tile_x', [-1, 32])],None,56
+ No: 4 GFLOPS: 1.62/11.79 result: MeasureResult(costs=(0.1653758006,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.756084442138672, timestamp=1656094760.3594854) [('tile_y', [-1, 1]), ('tile_x', [-1, 4])],None,20
+ No: 5 GFLOPS: 3.59/11.79 result: MeasureResult(costs=(0.07469718360000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.333705186843872, timestamp=1656094761.8218672) [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
+ No: 6 GFLOPS: 1.82/11.79 result: MeasureResult(costs=(0.14769602440000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.4847099781036377, timestamp=1656094764.8702123) [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
+ No: 7 GFLOPS: 0.87/11.79 result: MeasureResult(costs=(0.30947291220000006,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.071913480758667, timestamp=1656094769.9853463) [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
+ No: 8 GFLOPS: 10.29/11.79 result: MeasureResult(costs=(0.026079843599999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5585145950317383, timestamp=1656094770.5652766) [('tile_y', [-1, 4]), ('tile_x', [-1, 64])],None,62
+ No: 9 GFLOPS: 1.89/11.79 result: MeasureResult(costs=(0.141683723,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3673007488250732, timestamp=1656094773.0544333) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
+ No: 10 GFLOPS: 2.76/11.79 result: MeasureResult(costs=(0.0970883894,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6582880020141602, timestamp=1656094774.771685) [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 032c455b1..f5f6eb940 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -314,7 +314,7 @@ standard deviation.
.. code-block:: none
- {'mean': 497.09820047999983, 'median': 497.0073505500068, 'std': 0.594892070012249}
+ {'mean': 493.0292147700038, 'median': 493.25592685001993, 'std': 0.9379756005430333}
@@ -550,31 +550,31 @@ the tuning data to.
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
-
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 17.42/ 17.42 GFLOPS | Progress: (4/20) | 6.25 s
[Task 1/25] Current/Best: 6.17/ 17.42 GFLOPS | Progress: (8/20) | 9.22 s
[Task 1/25] Current/Best: 11.55/ 22.81 GFLOPS | Progress: (12/20) | 11.63 s
[Task 1/25] Current/Best: 16.86/ 22.81 GFLOPS | Progress: (16/20) | 13.31 s
[Task 1/25] Current/Best: 11.63/ 23.87 GFLOPS | Progress: (20/20) | 15.05 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.18/ 13.11 GFLOPS | Progress: (4/20) | 3.63 s
[Task 2/25] Current/Best: 13.96/ 18.31 GFLOPS | Progress: (8/20) | 4.93 s
[Task 2/25] Current/Best: 21.21/ 21.21 GFLOPS | Progress: (12/20) | 6.24 s
[Task 2/25] Current/Best: 11.69/ 21.21 GFLOPS | Progress: (16/20) | 7.54 s
[Task 2/25] Current/Best: 19.08/ 21.21 GFLOPS | Progress: (20/20) | 9.14 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 1.63/ 10.60 GFLOPS | Progress: (4/20) | 5.85 s
[Task 3/25] Current/Best: 15.51/ 16.83 GFLOPS | Progress: (8/20) | 7.78 s
[Task 3/25] Current/Best: 14.86/ 16.83 GFLOPS | Progress: (12/20) | 9.50 s
[Task 3/25] Current/Best: 7.19/ 23.81 GFLOPS | Progress: (16/20) | 11.45 s
[Task 3/25] Current/Best: 12.55/ 23.81 GFLOPS | Progress: (20/20) | 15.95 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 9.56/ 20.46 GFLOPS | Progress: (4/20) | 2.37 s
[Task 4/25] Current/Best: 6.75/ 20.46 GFLOPS | Progress: (8/20) | 6.71 s
[Task 4/25] Current/Best: 22.08/ 22.08 GFLOPS | Progress: (12/20) | 11.21 s
[Task 4/25] Current/Best: 17.35/ 22.08 GFLOPS | Progress: (16/20) | 13.45 s
[Task 4/25] Current/Best: 13.18/ 22.08 GFLOPS | Progress: (20/20) | 15.42 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 9.35/ 10.31 GFLOPS | Progress: (4/20) | 2.58 s
[Task 5/25] Current/Best: 11.53/ 12.40 GFLOPS | Progress: (8/20) | 4.66 s
[Task 5/25] Current/Best: 11.55/ 17.68 GFLOPS | Progress: (12/20) | 7.73 s
[Task 5/25] Current/Best: 11.69/ 22.86 GFLOPS | Progress: (16/20) | 9.14 s
[Task 5/25] Current/Best: 12.01/ 22.86 GFLOPS | Progress: (20/20) | 11.02 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 12.21/ 20.70 GFLOPS | Progress: (4/20) | 3.98 s
[Task 6/25] Current/Best: 18.85/ 20.70 GFLOPS | Progress: (8/20) | 5.74 s
[Task 6/25] Current/Best: 13.28/ 20.70 GFLOPS | Progress: (12/20) | 7.66 s
[Task 6/25] Current/Best: 19.84/ 20.70 GFLOPS | Progress: (16/20) | 9.94 s
[Task 6/25] Current/Best: 3.69/ 20.70 GFLOPS | Progress: (20/20) | 12.48 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 11.19/ 12.92 GFLOPS | Progress: (4/20) | 3.53 s
[Task 7/25] Current/Best: 20.31/ 21.04 GFLOPS | Progress: (8/20) | 5.03 s
[Task 7/25] Current/Best: 16.12/ 21.04 GFLOPS | Progress: (12/20) | 6.98 s
[Task 7/25] Current/Best: 12.19/ 21.04 GFLOPS | Progress: (16/20) | 9.01 s
[Task 7/25] Current/Best: 6.36/ 21.72 GFLOPS | Progress: (20/20) | 11.47 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 9.64/ 14.07 GFLOPS | Progress: (4/20) | 2.91 s
[Task 8/25] Current/Best: 9.42/ 14.07 GFLOPS | Progress: (8/20) | 7.75 s
[Task 8/25] Current/Best: 12.51/ 14.07 GFLOPS | Progress: (12/20) | 13.96 s
[Task 8/25] Current/Best: 18.40/ 18.40 GFLOPS | Progress: (16/20) | 16.05 s
[Task 8/25] Current/Best: 19.07/ 19.07 GFLOPS | Progress: (20/20) | 22.52 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 14.27/ 15.83 GFLOPS | Progress: (4/20) | 11.94 s
[Task 9/25] Current/Best: 23.53/ 23.53 GFLOPS | Progress: (8/20) | 13.74 s
[Task 9/25] Current/Best: 8.22/ 23.53 GFLOPS | Progress: (12/20) | 16.11 s
[Task 9/25] Current/Best: 17.89/ 23.53 GFLOPS | Progress: (16/20) | 18.77 s
[Task 9/25] Current/Best: 9.02/ 23.53 GFLOPS | Progress: (20/20) | 26.30 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 18.23/ 18.23 GFLOPS | Progress: (4/20) | 2.57 s
[Task 10/25] Current/Best: 15.53/ 18.23 GFLOPS | Progress: (8/20) | 4.14 s
[Task 10/25] Current/Best: 11.99/ 18.88 GFLOPS | Progress: (12/20) | 5.66 s
[Task 10/25] Current/Best: 18.60/ 20.50 GFLOPS | Progress: (16/20) | 6.77 s
[Task 10/25] Current/Best: 8.80/ 20.50 GFLOPS | Progress: (20/20
) | 8.32 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 11.13/ 18.14 GFLOPS | Progress: (4/20) | 3.33 s
[Task 11/25] Current/Best: 16.57/ 18.14 GFLOPS | Progress: (8/20) | 6.04 s
[Task 11/25] Current/Best: 18.18/ 18.18 GFLOPS | Progress: (12/20) | 8.07 s
[Task 11/25] Current/Best: 13.45/ 21.19 GFLOPS | Progress: (16/20) | 10.84 s
[Task 11/25] Current/Best: 19.45/ 21.41 GFLOPS | Progress: (20/20) | 12.89 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 7.79/ 18.03 GFLOPS | Progress: (4/20) | 5.44 s
[Task 12/25] Current/Best: 5.22/ 18.03 GFLOPS | Progress: (8/20) | 9.14 s
[Task 12/25] Current/Best: 18.93/ 18.98 GFLOPS | Progress: (12/20) | 11.14 s
[Task 12/25] Current/Best: 15.16/ 18.98 GFLOPS | Progress: (16/20) | 13.94 s
[Task 12/25] Current/Best: 15.12/ 19.06 GFLOPS | Progress: (20/20) | 15.86 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 8.65/ 17.37 GFLOPS | Progress: (4/20) | 3.68 s
[Task 13/25] Current/Best: 15.96/ 20.92 GFLOPS | Progress: (8/20) | 6.09 s
[Task 13/25] Current/Best: 19.62/ 21.66 GFLOPS | Progress: (12/20) | 8.95 s
[Task 13/25] Current/Best: 12.23/ 21.66 GFLOPS | Progress: (16/20) | 12.31 s
[Task 13/25] Current/Best: 18.57/ 21.66 GFLOPS | Progress: (20/20) | 14.53 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 13.65/ 13.65 GFLOPS | Progress: (4/20) | 3.28 s
[Task 14/25] Current/Best: 6.08/ 13.65 GFLOPS | Progress: (8/20) | 5.49 s
[Task 14/25] Current/Best: 20.77/ 20.77 GFLOPS | Progress: (12/20) | 8.01 s
[Task 14/25] Current/Best: 16.19/ 20.77 GFLOPS | Progress: (16/20) | 9.65 s Done.
-
[Task 14/25] Current/Best: 17.20/ 20.77 GFLOPS | Progress: (20/20) | 11.36 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 16.20/ 17.63 GFLOPS | Progress: (4/20) | 2.70 s
[Task 15/25] Current/Best: 14.45/ 18.13 GFLOPS | Progress: (8/20) | 4.03 s
[Task 15/25] Current/Best: 10.38/ 22.19 GFLOPS | Progress: (12/20) | 6.09 s
[Task 15/25] Current/Best: 20.40/ 22.19 GFLOPS | Progress: (16/20) | 9.42 s
[Task 15/25] Current/Best: 9.70/ 22.19 GFLOPS | Progress: (20/20) | 10.44 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 20.47/ 20.47 GFLOPS | Progress: (4/20) | 2.97 s
[Task 16/25] Current/Best: 3.04/ 20.47 GFLOPS | Progress: (8/20) | 4.58 s
[Task 16/25] Current/Best: 19.58/ 20.47 GFLOPS | Progress: (12/20) | 5.81 s
[Task 16/25] Current/Best: 18.06/ 20.47 GFLOPS | Progress: (16/20) |
7.15 s
[Task 16/25] Current/Best: 9.98/ 22.33 GFLOPS | Progress: (20/20) | 9.19 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 13.31/ 16.81 GFLOPS | Progress: (4/20) | 4.71 s
[Task 17/25] Current/Best: 14.39/ 23.25 GFLOPS | Progress: (8/20) | 7.55 s
[Task 17/25] Current/Best: 16.84/ 23.25 GFLOPS | Progress: (12/20) | 9.60 s
[Task 17/25] Current/Best: 16.41/ 23.25 GFLOPS | Progress: (16/20) | 11.75 s
[Task 17/25] Current/Best: 10.02/ 23.25 GFLOPS | Progress: (20/20) | 13.87 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 10.99/ 17.85 GFLOPS | Progress: (4/20) | 3.69 s
[Task 18/25] Current/Best: 10.55/ 17.85 GFLOPS | Progress: (8/20) | 7.22 s
[Task 18/25] Current/Best: 19.17/ 19.17 GFLOPS | Progress: (12/20) | 9.14 s
[Task 18/25] Current/Best: 9.95/ 19.17 GFLOPS | Progress: (16/20) | 12.68 s
[Task 18/25] Current/Best: 20.65/ 20.65 GFLOPS | Progress: (20/20) | 14.20 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 7.14/ 20.26 GFLOPS | Progress: (4/20) | 6.10 s
[Task 19/25] Current/Best: 2.61/ 20.26 GFLOPS | Progress: (8/20) | 9.36 s
[Task 19/25] Current/Best: 19.71/ 21.86 GFLOPS | Progress: (12/20) | 12.09 s
[Task 19/25] Current/Best: 15.07/ 22.26 GFLOPS | Progress: (16/20) | 14.91 s
[Task 19/25] Current/Best: 2.70/ 23.70 GFLOPS | Progress: (20/20) | 17.72 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 9.24/ 15.13 GFLOPS | Progress: (4/20) | 3.32 s Done.
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 17.52/ 17.52 GFLOPS | Progress: (4/20) | 6.26 s
[Task 1/25] Current/Best: 6.15/ 17.52 GFLOPS | Progress: (8/20) | 9.18 s
[Task 1/25] Current/Best: 11.54/ 22.78 GFLOPS | Progress: (12/20) | 11.61 s
[Task 1/25] Current/Best: 16.82/ 22.78 GFLOPS | Progress: (16/20) | 13.30 s
[Task 1/25] Current/Best: 11.59/ 23.70 GFLOPS | Progress: (20/20) | 15.05 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.21/ 13.20 GFLOPS | Progress: (4/20) | 3.65 s
[Task 2/25] Current/Best: 14.09/ 18.78 GFLOPS | Progress: (8/20) | 4.93 s
[Task 2/25] Current/Best: 21.32/ 21.32 GFLOPS | Progress: (12/20) | 6.25 s
[Task 2/25] Current/Best: 12.45/ 21.32 GFLOPS | Progress: (16/20) | 7.51 s
[Task 2/25] Current/Best: 19.12/ 21.32 GFLOPS | Progress: (20/20) | 9.12 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 1.63/ 10.56 GFLOPS | Progress: (4/20) | 5.87 s
[Task 3/25] Current/Best: 15.54/ 16.83 GFLOPS | Progress: (8/20) | 7.78 s
[Task 3/25] Current/Best: 14.89/ 16.83 GFLOPS | Progress: (12/20) | 9.49 s
[Task 3/25] Current/Best: 7.17/ 23.86 GFLOPS | Progress: (16/20) | 11.43 s
[Task 3/25] Current/Best: 12.60/ 23.86 GFLOPS | Progress: (20/20) | 15.99 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 9.57/ 20.55 GFLOPS | Progress: (4/20) | 2.39 s
[Task 4/25] Current/Best: 6.73/ 20.55 GFLOPS | Progress: (8/20) | 6.76 s
[Task 4/25] Current/Best: 22.00/ 22.00 GFLOPS | Progress: (12/20) | 11.32 s
[Task 4/25] Current/Best: 16.40/ 22.00 GFLOPS | Progress: (16/20) | 13.58 s
[Task 4/25] Current/Best: 13.42/ 22.00 GFLOPS | Progress: (20/20) | 15.57 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 9.48/ 10.14 GFLOPS | Progress: (4/20) | 2.57 s
[Task 5/25] Current/Best: 11.70/ 12.38 GFLOPS | Progress: (8/20) | 4.63 s
[Task 5/25] Current/Best: 10.95/ 18.04 GFLOPS | Progress: (12/20) | 7.73 s
[Task 5/25] Current/Best: 11.75/ 22.43 GFLOPS | Progress: (16/20) | 9.15 s
[Task 5/25] Current/Best: 12.07/ 22.43 GFLOPS | Progress: (20/20) | 11.00 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 12.30/ 20.67 GFLOPS | Progress: (4/20) | 3.98 s
[Task 6/25] Current/Best: 19.01/ 20.67 GFLOPS | Progress: (8/20) | 5.73 s
[Task 6/25] Current/Best: 13.26/ 20.67 GFLOPS | Progress: (12/20) | 7.65 s
[Task 6/25] Current/Best: 20.01/ 20.67 GFLOPS | Progress: (16/20) | 9.89 s
[Task 6/25] Current/Best: 3.73/ 20.67 GFLOPS | Progress: (20/20) | 12.39 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 11.21/ 12.79 GFLOPS | Progress: (4/20) | 3.54 s
[Task 7/25] Current/Best: 20.18/ 21.11 GFLOPS | Progress: (8/20) | 5.04 s
[Task 7/25] Current/Best: 16.09/ 21.11 GFLOPS | Progress: (12/20) | 6.93 s
[Task 7/25] Current/Best: 12.22/ 21.11 GFLOPS | Progress: (16/20) | 8.98 s
[Task 7/25] Current/Best: 6.37/ 21.63 GFLOPS | Progress: (20/20) | 11.43 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 9.96/ 14.13 GFLOPS | Progress: (4/20) | 2.93 s
[Task 8/25] Current/Best: 9.47/ 14.13 GFLOPS | Progress: (8/20) | 7.72 s
[Task 8/25] Current/Best: 13.19/ 14.13 GFLOPS | Progress: (12/20) | 13.83 s
[Task 8/25] Current/Best: 18.48/ 18.48 GFLOPS | Progress: (16/20) | 15.93 s
[Task 8/25] Current/Best: 19.67/ 19.67 GFLOPS | Progress: (20/20) | 22.52 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 14.37/ 15.79 GFLOPS | Progress: (4/20) | 11.94 s
[Task 9/25] Current/Best: 23.34/ 23.34 GFLOPS | Progress: (8/20) | 13.72 s
[Task 9/25] Current/Best: 8.23/ 23.34 GFLOPS | Progress: (12/20) | 16.05 s
[Task 9/25] Current/Best: 17.93/ 23.34 GFLOPS | Progress: (16/20) | 18.77 s
[Task 9/25] Current/Best: 9.03/ 23.34 GFLOPS | Progress: (20/20) | 26.49 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 18.18/ 18.18 GFLOPS | Progress: (4/20) | 2.58 s
[Task 10/25] Current/Best: 15.54/ 18.18 GFLOPS | Progress: (8/20) | 4.17 s
[Task 10/25] Current/Best: 12.63/ 18.94 GFLOPS | Progress: (12/20) | 5.69 s
[Task 10/25] Current/Best: 19.10/ 20.14 GFLOPS | Progress: (16/20) | 6.80 s
[Task 10/25] Current/Best: 8.83/ 20.14 GFLOPS | Progress: (20/20
) | 8.33 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 12.35/ 18.15 GFLOPS | Progress: (4/20) | 3.34 s
[Task 11/25] Current/Best: 16.95/ 18.15 GFLOPS | Progress: (8/20) | 6.09 s
[Task 11/25] Current/Best: 18.22/ 18.22 GFLOPS | Progress: (12/20) | 8.11 s
[Task 11/25] Current/Best: 13.22/ 21.18 GFLOPS | Progress: (16/20) | 10.87 s
[Task 11/25] Current/Best: 19.44/ 21.57 GFLOPS | Progress: (20/20) | 12.89 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 7.79/ 18.07 GFLOPS | Progress: (4/20) | 5.31 s
[Task 12/25] Current/Best: 5.28/ 18.07 GFLOPS | Progress: (8/20) | 9.04 s
[Task 12/25] Current/Best: 18.83/ 18.97 GFLOPS | Progress: (12/20) | 11.02 s
[Task 12/25] Current/Best: 15.37/ 18.97 GFLOPS | Progress: (16/20) | 13.84 s
[Task 12/25] Current/Best: 15.14/ 18.97 GFLOPS | Progress: (20/20) | 15.76 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 8.73/ 17.31 GFLOPS | Progress: (4/20) | 3.64 s
[Task 13/25] Current/Best: 16.10/ 21.05 GFLOPS | Progress: (8/20) | 6.09 s
[Task 13/25] Current/Best: 19.57/ 21.63 GFLOPS | Progress: (12/20) | 9.00 s
[Task 13/25] Current/Best: 12.24/ 21.63 GFLOPS | Progress: (16/20) | 12.42 s
[Task 13/25] Current/Best: 18.80/ 21.63 GFLOPS | Progress: (20/20) | 14.67 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 13.61/ 13.61 GFLOPS | Progress: (4/20) | 3.32 s
[Task 14/25] Current/Best: 6.07/ 13.61 GFLOPS | Progress: (8/20) | 5.49 s
[Task 14/25] Current/Best: 19.96/ 19.96 GFLOPS | Progress: (12/20) | 8.06 s
[Task 14/25] Current/Best: 17.08/ 19.96 GFLOPS | Progress: (16/20) | 9.69 s Done.
+
[Task 14/25] Current/Best: 17.25/ 19.96 GFLOPS | Progress: (20/20) | 11.42 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 16.18/ 17.66 GFLOPS | Progress: (4/20) | 2.69 s
[Task 15/25] Current/Best: 14.42/ 18.12 GFLOPS | Progress: (8/20) | 4.02 s
[Task 15/25] Current/Best: 10.40/ 22.00 GFLOPS | Progress: (12/20) | 6.08 s
[Task 15/25] Current/Best: 19.88/ 22.00 GFLOPS | Progress: (16/20) | 9.54 s
[Task 15/25] Current/Best: 9.68/ 22.00 GFLOPS | Progress: (20/20) | 10.55 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 20.36/ 20.36 GFLOPS | Progress: (4/20) | 3.10 s
[Task 16/25] Current/Best: 3.01/ 20.36 GFLOPS | Progress: (8/20) | 4.73 s
[Task 16/25] Current/Best: 19.18/ 20.36 GFLOPS | Progress: (12/20) | 5.97 s
[Task 16/25] Current/Best: 17.37/ 20.36 GFLOPS | Progress: (16/20) |
7.36 s
[Task 16/25] Current/Best: 10.05/ 20.36 GFLOPS | Progress: (20/20) | 9.43 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 11.81/ 18.49 GFLOPS | Progress: (4/20) | 4.72 s
[Task 17/25] Current/Best: 13.99/ 23.35 GFLOPS | Progress: (8/20) | 7.57 s
[Task 17/25] Current/Best: 17.46/ 23.35 GFLOPS | Progress: (12/20) | 9.62 s
[Task 17/25] Current/Best: 16.51/ 23.35 GFLOPS | Progress: (16/20) | 11.77 s
[Task 17/25] Current/Best: 10.03/ 23.35 GFLOPS | Progress: (20/20) | 13.88 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 11.44/ 17.88 GFLOPS | Progress: (4/20) | 3.68 s
[Task 18/25] Current/Best: 10.71/ 17.88 GFLOPS | Progress: (8/20) | 7.15 s
[Task 18/25] Current/Best: 19.14/ 19.14 GFLOPS | Progress: (12/20) | 9.06 s
[Task 18/25] Current/Best: 10.10/ 19.14 GFLOPS | Progress: (16/20) | 12.60 s
[Task 18/25] Current/Best: 20.55/ 20.55 GFLOPS | Progress: (20/20) | 14.10 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 7.16/ 20.18 GFLOPS | Progress: (4/20) | 5.99 s
[Task 19/25] Current/Best: 2.61/ 20.18 GFLOPS | Progress: (8/20) | 9.26 s
[Task 19/25] Current/Best: 19.58/ 21.71 GFLOPS | Progress: (12/20) | 12.10 s
[Task 19/25] Current/Best: 14.47/ 21.89 GFLOPS | Progress: (16/20) | 14.97 s
[Task 19/25] Current/Best: 2.70/ 23.46 GFLOPS | Progress: (20/20) | 17.79 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 8.98/ 15.28 GFLOPS | Progress: (4/20) | 3.28 s Done.
Done.
-
[Task 20/25] Current/Best: 10.40/ 15.13 GFLOPS | Progress: (8/20) | 6.76 s
[Task 20/25] Current/Best: 2.32/ 16.65 GFLOPS | Progress: (12/20) | 10.61 s
[Task 20/25] Current/Best: 12.40/ 16.65 GFLOPS | Progress: (16/20) | 14.15 s
[Task 20/25] Current/Best: 11.53/ 22.15 GFLOPS | Progress: (20/20) | 16.27 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 6.42/ 17.70 GFLOPS | Progress: (4/20) | 3.23 s
[Task 21/25] Current/Best: 14.64/ 17.70 GFLOPS | Progress: (8/20) | 4.80 s
[Task 21/25] Current/Best: 1.61/ 17.70 GFLOPS | Progress: (12/20) | 6.95 s
[Task 21/25] Current/Best: 17.89/ 17.89 GFLOPS | Progress: (16/20) | 10.38 s
[Task 21/25] Current/Best: 4.47/ 17.89 GFLOPS | Progress: (20/20) | 17.48 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 2.70/ 17.07 GFLOPS | Progress: (4/20
) | 2.66 s
[Task 22/25] Current/Best: 8.66/ 21.79 GFLOPS | Progress: (8/20) | 4.65 s
[Task 22/25] Current/Best: 19.98/ 21.79 GFLOPS | Progress: (12/20) | 6.94 s
[Task 22/25] Current/Best: 15.47/ 21.79 GFLOPS | Progress: (16/20) | 8.98 s
[Task 22/25] Current/Best: 14.57/ 21.79 GFLOPS | Progress: (20/20) | 10.63 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 17.51/ 20.34 GFLOPS | Progress: (4/20) | 3.25 s
[Task 23/25] Current/Best: 15.03/ 20.34 GFLOPS | Progress: (8/20) | 6.59 s
[Task 23/25] Current/Best: 20.97/ 21.45 GFLOPS | Progress: (12/20) | 8.38 s
[Task 23/25] Current/Best: 6.41/ 21.45 GFLOPS | Progress: (16/20) | 15.40 s
[Task 23/25] Current/Best: 7.86/ 21.45 GFLOPS | Progress: (20/20) | 19.58 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.47/ 8.47 GFLOPS | Progress: (4/20) | 11.79 s
[Task 24/25] Current/Best: 3.51/ 8.47 GFLOPS | Progress: (8/20) | 23.02 s
[Task 24/25] Current/Best: 4.24/ 8.47 GFLOPS | Progress: (12/20) | 33.77 s Done.
+
[Task 20/25] Current/Best: 9.69/ 15.28 GFLOPS | Progress: (8/20) | 6.57 s
[Task 20/25] Current/Best: 2.32/ 15.28 GFLOPS | Progress: (12/20) | 10.41 s
[Task 20/25] Current/Best: 12.25/ 15.28 GFLOPS | Progress: (16/20) | 13.98 s
[Task 20/25] Current/Best: 12.62/ 22.14 GFLOPS | Progress: (20/20) | 16.08 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 6.41/ 17.68 GFLOPS | Progress: (4/20) | 3.19 s
[Task 21/25] Current/Best: 14.64/ 17.68 GFLOPS | Progress: (8/20) | 4.79 s
[Task 21/25] Current/Best: 1.61/ 17.68 GFLOPS | Progress: (12/20) | 6.91 s
[Task 21/25] Current/Best: 15.94/ 17.68 GFLOPS | Progress: (16/20) | 10.35 s
[Task 21/25] Current/Best: 4.48/ 17.68 GFLOPS | Progress: (20/20) | 17.49 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 2.70/ 17.04 GFLOPS | Progress: (4/20
) | 2.66 s
[Task 22/25] Current/Best: 8.89/ 21.74 GFLOPS | Progress: (8/20) | 4.65 s
[Task 22/25] Current/Best: 19.94/ 21.74 GFLOPS | Progress: (12/20) | 6.97 s
[Task 22/25] Current/Best: 15.62/ 21.74 GFLOPS | Progress: (16/20) | 9.05 s
[Task 22/25] Current/Best: 13.75/ 21.74 GFLOPS | Progress: (20/20) | 10.77 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 17.60/ 20.76 GFLOPS | Progress: (4/20) | 3.21 s
[Task 23/25] Current/Best: 15.35/ 20.76 GFLOPS | Progress: (8/20) | 6.55 s
[Task 23/25] Current/Best: 19.91/ 21.63 GFLOPS | Progress: (12/20) | 8.36 s
[Task 23/25] Current/Best: 6.35/ 21.63 GFLOPS | Progress: (16/20) | 15.26 s
[Task 23/25] Current/Best: 7.95/ 21.63 GFLOPS | Progress: (20/20) | 19.46 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.25/ 8.25 GFLOPS | Progress: (4/20) | 11.79 s
[Task 24/25] Current/Best: 2.14/ 8.25 GFLOPS | Progress: (8/20) | 22.78 s
[Task 24/25] Current/Best: 4.25/ 8.25 GFLOPS | Progress: (12/20) | 34.30 s Done.
Done.
-
[Task 24/25] Current/Best: 6.15/ 8.93 GFLOPS | Progress: (16/20) | 39.22 s
[Task 24/25] Current/Best: 3.35/ 8.93 GFLOPS | Progress: (20/20) | 45.21 s Done.
-
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 25/25] Current/Best: 1.55/ 2.75 GFLOPS | Progress: (4/20) | 11.58 s
[Task 25/25] Current/Best: 5.85/ 8.14 GFLOPS | Progress: (8/20) | 22.83 s
[Task 25/25] Current/Best: 6.00/ 8.14 GFLOPS | Progress: (12/20) | 34.11 s
[Task 25/25] Current/Best: 5.82/ 8.35 GFLOPS | Progress: (16/20) | 35.85 s
[Task 25/25] Current/Best: 2.84/ 8.72 GFLOPS | Progress: (20/20) | 46.52 s
+
[Task 24/25] Current/Best: 6.05/ 8.77 GFLOPS | Progress: (16/20) | 39.74 s
[Task 24/25] Current/Best: 3.32/ 8.80 GFLOPS | Progress: (20/20) | 45.61 s Done.
+
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 25/25] Current/Best: 1.55/ 2.91 GFLOPS | Progress: (4/20) | 11.59 s
[Task 25/25] Current/Best: 5.53/ 7.97 GFLOPS | Progress: (8/20) | 22.87 s
[Task 25/25] Current/Best: 5.93/ 7.97 GFLOPS | Progress: (12/20) | 34.32 s
[Task 25/25] Current/Best: 5.78/ 8.46 GFLOPS | Progress: (16/20) | 36.21 s
[Task 25/25] Current/Best: 2.85/ 8.71 GFLOPS | Progress: (20/20) | 46.86 s
@@ -677,8 +677,8 @@ Verify that the optimized model runs and produces the same results:
.. code-block:: none
- class='n02123045 tabby, tabby cat' with probability=0.621105
- class='n02123159 tiger cat' with probability=0.356377
+ class='n02123045 tabby, tabby cat' with probability=0.621104
+ 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
class='n04040759 radiator' with probability=0.000262
@@ -735,8 +735,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 413.1419657599963, 'median': 413.0868064499964, 'std': 0.6840461854077801}
- unoptimized: {'mean': 497.09820047999983, 'median': 497.0073505500068, 'std': 0.594892070012249}
+ optimized: {'mean': 409.4395423199876, 'median': 409.0542080499745, 'std': 1.4722725743014862}
+ unoptimized: {'mean': 493.0292147700038, 'median': 493.25592685001993, 'std': 0.9379756005430333}
@@ -759,7 +759,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 10 minutes 13.278 seconds)
+ **Total running time of the script:** ( 10 minutes 13.444 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 6518d76b6..d148ba28c 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -269,7 +269,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.3e-07 secs/op
+ 1.284e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index b7e94b5c5..56e5f8d2a 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -262,7 +262,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0x21541620)), stage(b, placeholder(b, 0x22863260)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(mi [...]
+ [stage(a, placeholder(a, 0x1a89e6b0)), stage(b, placeholder(b, 0x21b69e00)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(mi [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index f06cd090f..968dbdc23 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,29 +5,29 @@
Computation times
=================
-**12:56.859** total execution time for **tutorial** files:
+**12:57.579** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 10:13.278 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 10:13.444 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 00:59.974 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 00:59.932 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 00:49.089 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 00:50.153 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:28.035 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:28.484 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:25.139 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:24.232 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.676 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.670 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.515 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.514 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.152 | 0.0 MB |
-+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``) | 00:00.000 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.149 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``) | 00:00.000 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``) | 00:00.000 | 0.0 MB |
++------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``) | 00:00.000 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_install.py` (``install.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 3f1788c78..9a4cac633 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -288,8 +288,8 @@ helper function to run a profile of the TVM generated code.
.. code-block:: none
- Numpy running time: 0.000010
- naive: 0.000006
+ Numpy running time: 0.000008
+ naive: 0.000007
@@ -390,7 +390,7 @@ compile and run this new schedule with the parallel operation applied:
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- parallel: 0.000006
+ parallel: 0.000007
@@ -499,10 +499,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 9.969230000024254e-06 1.0
- naive 5.8628e-06 0.588089551548689
- parallel 6.1027e-06 0.6121535966153006
- vector 2.4564299999999998e-05 2.4640117641924437
+ numpy 7.862000002205605e-06 1.0
+ naive 6.7153e-06 0.8541465273614971
+ parallel 7.261900000000001e-06 0.9236708214147479
+ vector 2.4612700000000002e-05 3.1305901797373643
@@ -983,7 +983,7 @@ optimizations.
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- none: 3.338866
+ none: 3.331012
@@ -1088,7 +1088,7 @@ schedule.
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- blocking: 0.304134
+ blocking: 0.304537
@@ -1186,7 +1186,7 @@ already cache friendly from our previous optimizations.
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- vectorization: 0.339931
+ vectorization: 0.341623
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1262,7 +1262,7 @@ more cache friendly.
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- loop permutation: 0.114518
+ loop permutation: 0.117015
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1363,7 +1363,7 @@ optimized schedule.
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- array packing: 0.108645
+ array packing: 0.108857
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1458,7 +1458,7 @@ to `C` when all the block results are ready.
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- block caching: 0.111016
+ block caching: 0.110480
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1546,7 +1546,7 @@ of thread-level parallelization.
/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- parallelization: 0.144471
+ parallelization: 0.144206
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1627,13 +1627,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.3388662441999997 1.0
- blocking 0.30413420599999996 0.09108906549590495
- vectorization 0.3399305729 0.10181017987482999
- loop permutation 0.1145181479 0.03429851318510629
- array packing 0.108644995 0.03253948707550925
- block caching 0.1110158872 0.03324957607776219
- parallelization 0.1444712218 0.043269544579979315
+ none 3.3310115866 1.0
+ blocking 0.3045366174 0.0914246646949805
+ vectorization 0.34162335580000003 0.10255844115772011
+ loop permutation 0.11701485920000002 0.035128925900686635
+ array packing 0.10885721129999999 0.03267992574325199
+ block caching 0.11048004900000001 0.03316711639324203
+ parallelization 0.14420581970000002 0.04329189975805292
diff --git a/docs/commit_hash b/docs/commit_hash
index fe90e12c8..23e8f0831 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-f1d30a27b2efe5b15e6492f785be1d41c9a75ab9
+2c3b56c9d3fa07407eb6dda7f2142f6ba3748ecf
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 29109cd4e..6d0c6cbf2 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -422,7 +422,7 @@ to download the full example code</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"x"</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">x</span><span class="o">.</span><span class="n">shape</span></a><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip7711d7c2-bb2d-4ab9-b495-a12f8fe83ee8 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.zip5651072b-8a77-4627-9e37-9128756bd30b 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 91c1a3cb6..a1383714c 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -427,41 +427,45 @@ python3 -m pip install -f https://release.oneflow.info <span class="nv">oneflow<
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
0%| | 0.00/41.5M [00:00<?, ?B/s]
- 0%| | 16.0k/41.5M [00:00<07:42, 94.0kB/s]
- 0%| | 48.0k/41.5M [00:00<04:52, 149kB/s]
- 0%| | 96.0k/41.5M [00:00<03:27, 209kB/s]
- 0%| | 208k/41.5M [00:00<01:52, 385kB/s]
- 1%|1 | 432k/41.5M [00:00<00:59, 718kB/s]
- 2%|2 | 872k/41.5M [00:01<00:31, 1.35MB/s]
- 4%|4 | 1.72M/41.5M [00:01<00:16, 2.60MB/s]
- 8%|7 | 3.26M/41.5M [00:01<00:08, 4.70MB/s]
- 12%|#1 | 4.80M/41.5M [00:01<00:06, 6.11MB/s]
- 15%|#5 | 6.34M/41.5M [00:01<00:05, 7.07MB/s]
- 19%|#8 | 7.88M/41.5M [00:01<00:04, 7.72MB/s]
- 23%|##2 | 9.41M/41.5M [00:02<00:04, 8.17MB/s]
- 26%|##6 | 10.9M/41.5M [00:02<00:03, 8.47MB/s]
- 30%|### | 12.5M/41.5M [00:02<00:03, 8.69MB/s]
- 34%|###3 | 14.0M/41.5M [00:02<00:03, 8.82MB/s]
- 37%|###7 | 15.5M/41.5M [00:02<00:03, 8.93MB/s]
- 41%|####1 | 17.1M/41.5M [00:02<00:02, 9.02MB/s]
- 45%|####4 | 18.6M/41.5M [00:03<00:02, 9.06MB/s]
- 49%|####8 | 20.2M/41.5M [00:03<00:02, 9.10MB/s]
- 52%|#####2 | 21.7M/41.5M [00:03<00:02, 9.09MB/s]
- 56%|#####5 | 23.2M/41.5M [00:03<00:02, 9.12MB/s]
- 60%|#####9 | 24.8M/41.5M [00:03<00:01, 9.15MB/s]
- 63%|######3 | 26.3M/41.5M [00:04<00:01, 9.16MB/s]
- 66%|######5 | 27.2M/41.5M [00:04<00:01, 7.56MB/s]
- 69%|######9 | 28.8M/41.5M [00:04<00:01, 8.03MB/s]
- 71%|#######1 | 29.6M/41.5M [00:04<00:01, 7.69MB/s]
- 75%|#######5 | 31.2M/41.5M [00:04<00:01, 8.16MB/s]
- 79%|#######8 | 32.7M/41.5M [00:04<00:01, 8.47MB/s]
- 83%|########2 | 34.2M/41.5M [00:05<00:00, 8.70MB/s]
- 86%|########6 | 35.8M/41.5M [00:05<00:00, 8.84MB/s]
- 90%|########9 | 37.3M/41.5M [00:05<00:00, 8.95MB/s]
- 94%|#########3| 38.9M/41.5M [00:05<00:00, 9.00MB/s]
- 97%|#########7| 40.4M/41.5M [00:05<00:00, 9.06MB/s]
- 99%|#########9| 41.3M/41.5M [00:06<00:00, 7.35MB/s]
-100%|##########| 41.5M/41.5M [00:06<00:00, 7.24MB/s]
+ 0%| | 16.0k/41.5M [00:00<08:06, 89.4kB/s]
+ 0%| | 40.0k/41.5M [00:00<06:16, 115kB/s]
+ 0%| | 96.0k/41.5M [00:00<03:31, 205kB/s]
+ 0%| | 168k/41.5M [00:00<02:33, 282kB/s]
+ 1%| | 352k/41.5M [00:00<01:18, 549kB/s]
+ 2%|1 | 664k/41.5M [00:01<00:44, 952kB/s]
+ 3%|3 | 1.30M/41.5M [00:01<00:22, 1.85MB/s]
+ 6%|6 | 2.59M/41.5M [00:01<00:11, 3.61MB/s]
+ 10%|9 | 4.06M/41.5M [00:01<00:07, 5.09MB/s]
+ 13%|#3 | 5.53M/41.5M [00:01<00:06, 6.10MB/s]
+ 17%|#6 | 7.00M/41.5M [00:02<00:05, 6.79MB/s]
+ 20%|## | 8.46M/41.5M [00:02<00:04, 7.26MB/s]
+ 24%|##3 | 9.93M/41.5M [00:02<00:04, 7.59MB/s]
+ 27%|##7 | 11.4M/41.5M [00:02<00:04, 7.84MB/s]
+ 31%|###1 | 12.9M/41.5M [00:02<00:03, 7.99MB/s]
+ 35%|###4 | 14.3M/41.5M [00:02<00:03, 8.09MB/s]
+ 38%|###8 | 15.8M/41.5M [00:03<00:03, 8.17MB/s]
+ 42%|####1 | 17.3M/41.5M [00:03<00:03, 8.23MB/s]
+ 45%|####5 | 18.7M/41.5M [00:03<00:02, 8.27MB/s]
+ 49%|####8 | 20.2M/41.5M [00:03<00:02, 8.30MB/s]
+ 52%|#####2 | 21.7M/41.5M [00:03<00:02, 8.31MB/s]
+ 56%|#####5 | 23.1M/41.5M [00:04<00:02, 8.33MB/s]
+ 59%|#####9 | 24.6M/41.5M [00:04<00:02, 8.34MB/s]
+ 63%|######2 | 26.1M/41.5M [00:04<00:01, 8.33MB/s]
+ 66%|######6 | 27.5M/41.5M [00:04<00:01, 9.17MB/s]
+ 69%|######9 | 28.8M/41.5M [00:04<00:01, 9.97MB/s]
+ 72%|#######1 | 29.8M/41.5M [00:04<00:01, 8.98MB/s]
+ 74%|#######4 | 30.7M/41.5M [00:04<00:01, 7.84MB/s]
+ 77%|#######6 | 31.9M/41.5M [00:05<00:01, 8.49MB/s]
+ 80%|#######9 | 33.1M/41.5M [00:05<00:00, 9.36MB/s]
+ 82%|########2 | 34.1M/41.5M [00:05<00:00, 8.39MB/s]
+ 84%|########4 | 34.9M/41.5M [00:05<00:00, 7.31MB/s]
+ 88%|########7 | 36.3M/41.5M [00:05<00:00, 8.48MB/s]
+ 90%|######### | 37.5M/41.5M [00:05<00:00, 9.28MB/s]
+ 93%|#########2| 38.4M/41.5M [00:05<00:00, 8.31MB/s]
+ 95%|#########4| 39.3M/41.5M [00:06<00:00, 8.19MB/s]
+ 97%|#########7| 40.4M/41.5M [00:06<00:00, 9.07MB/s]
+100%|#########9| 41.3M/41.5M [00:06<00:00, 8.04MB/s]
+100%|##########| 41.5M/41.5M [00:06<00:00, 6.93MB/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 16736c7cb..b8b8718c3 100644
--- a/docs/how_to/compile_models/from_paddle.html
+++ b/docs/how_to/compile_models/from_paddle.html
@@ -488,7 +488,7 @@ A quick solution is</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>TVM prediction top-1 id: 282, class name: 282: 'tiger cat',
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 5.825 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 6.676 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 3bda207b8..14c9e3bee 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -409,10 +409,12 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
0%| | 0.00/44.7M [00:00<?, ?B/s]
- 9%|8 | 3.95M/44.7M [00:00<00:01, 41.4MB/s]
- 27%|##7 | 12.2M/44.7M [00:00<00:00, 68.1MB/s]
- 62%|######2 | 27.8M/44.7M [00:00<00:00, 111MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 122MB/s]
+ 5%|4 | 2.07M/44.7M [00:00<00:02, 21.6MB/s]
+ 13%|#3 | 5.88M/44.7M [00:00<00:01, 32.2MB/s]
+ 29%|##9 | 13.0M/44.7M [00:00<00:00, 51.7MB/s]
+ 59%|#####8 | 26.3M/44.7M [00:00<00:00, 85.9MB/s]
+ 82%|########1 | 36.4M/44.7M [00:00<00:00, 93.3MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 79.5MB/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 f6f3dbd20..240829d50 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -631,7 +631,7 @@ banana (score = 0.00022)
desk (score = 0.00019)
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 1.946 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 1.995 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 81ae3421b..3f5dc0810 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -322,7 +322,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:22.379</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:40.658</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -331,43 +331,43 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="from_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>01:05.825</p></td>
+<td><p>01:06.676</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><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>01:01.946</p></td>
+<td><p>01:01.995</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><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>00:57.694</p></td>
+<td><p>00:59.175</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>00:31.859</p></td>
+<tr class="row-even"><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>00:39.645</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>00:23.901</p></td>
+<tr class="row-odd"><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>00:32.333</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>00:23.763</p></td>
+<td><p>00:23.262</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></td>
-<td><p>00:21.567</p></td>
+<td><p>00:21.589</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></td>
-<td><p>00:19.344</p></td>
+<td><p>00:19.486</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_keras.html#sphx-glr-how-to-compile-models-from-keras-py"><span class="std std-ref">Compile Keras Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_keras.py</span></code>)</p></td>
-<td><p>00:14.126</p></td>
+<td><p>00:14.102</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>00:02.355</p></td>
+<td><p>00:02.395</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 2bceed688..9ebcca174 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -648,7 +648,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.9979 15.9347 16.5312 15.7462 0.2189
+ 16.0560 16.0311 16.3316 15.9755 0.0957
</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 f4dd8492d..f12da9cda 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -431,17 +431,19 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
0%| | 0.00/170M [00:00<?, ?B/s]
- 1%| | 1.27M/170M [00:00<00:13, 13.1MB/s]
- 2%|2 | 3.42M/170M [00:00<00:09, 18.4MB/s]
- 5%|4 | 8.40M/170M [00:00<00:05, 33.6MB/s]
- 11%|#1 | 19.1M/170M [00:00<00:02, 64.2MB/s]
- 24%|##3 | 40.3M/170M [00:00<00:01, 121MB/s]
- 37%|###7 | 63.6M/170M [00:00<00:00, 163MB/s]
- 51%|#####1 | 87.3M/170M [00:00<00:00, 191MB/s]
- 65%|######5 | 111M/170M [00:00<00:00, 208MB/s]
- 79%|#######9 | 135M/170M [00:00<00:00, 221MB/s]
- 93%|#########3| 158M/170M [00:01<00:00, 229MB/s]
-100%|##########| 170M/170M [00:01<00:00, 169MB/s]
+ 1%| | 944k/170M [00:00<00:18, 9.59MB/s]
+ 2%|1 | 2.64M/170M [00:00<00:12, 14.5MB/s]
+ 3%|3 | 5.83M/170M [00:00<00:07, 23.0MB/s]
+ 7%|6 | 11.7M/170M [00:00<00:04, 38.2MB/s]
+ 14%|#3 | 23.0M/170M [00:00<00:02, 66.8MB/s]
+ 22%|##1 | 37.1M/170M [00:00<00:01, 94.6MB/s]
+ 33%|###3 | 56.4M/170M [00:00<00:00, 130MB/s]
+ 45%|####4 | 76.2M/170M [00:00<00:00, 154MB/s]
+ 56%|#####6 | 95.7M/170M [00:00<00:00, 170MB/s]
+ 68%|######7 | 115M/170M [00:01<00:00, 181MB/s]
+ 80%|#######9 | 135M/170M [00:01<00:00, 189MB/s]
+ 91%|#########1| 155M/170M [00:01<00:00, 194MB/s]
+100%|##########| 170M/170M [00:01<00:00, 138MB/s]
/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:3878: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/usr/local/lib/python3.7/dist-packages/torchvision/models/detection/anchor_utils.py:127: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -536,7 +538,7 @@ torchvision rcnn models.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Get 9 valid boxes
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 54.429 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 1.189 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 338900983..ea056875f 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -472,11 +472,10 @@ training. Other models require a full post training calibration.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
0%| | 0.00/13.6M [00:00<?, ?B/s]
- 6%|6 | 880k/13.6M [00:00<00:01, 8.96MB/s]
- 18%|#8 | 2.48M/13.6M [00:00<00:00, 13.6MB/s]
- 40%|#### | 5.43M/13.6M [00:00<00:00, 21.5MB/s]
- 80%|#######9 | 10.8M/13.6M [00:00<00:00, 35.2MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 32.6MB/s]
+ 9%|9 | 1.23M/13.6M [00:00<00:01, 12.7MB/s]
+ 27%|##7 | 3.73M/13.6M [00:00<00:00, 20.5MB/s]
+ 65%|######4 | 8.75M/13.6M [00:00<00:00, 35.1MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 39.1MB/s]
</pre></div>
</div>
</div>
@@ -565,7 +564,7 @@ output values are identical out of 1000 outputs from mobilenet v2.</p>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.4646 90.2163 100.8701 90.0853 1.3153
+ 90.3201 90.2930 91.0299 90.1665 0.1299
</pre></div>
</div>
<div class="admonition note">
@@ -604,7 +603,7 @@ This includes support for the VNNI 8 bit dot product instruction (CascadeLake or
<div class="section" id="deploy-a-quantized-tflite-model">
<h2>Deploy a quantized TFLite Model<a class="headerlink" href="#deploy-a-quantized-tflite-model" title="Permalink to this headline">¶</a></h2>
<p>TODO</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 7.399 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 8.754 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 c06f90463..272c1dba0 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -565,7 +565,7 @@ 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)
- 119.3751 119.3254 120.2596 118.6613 0.3022
+ 120.2101 120.1123 125.5592 119.5570 0.6454
</pre></div>
</div>
<div class="admonition note">
@@ -593,7 +593,7 @@ network for ARM CPU</span></a>.</p></li>
</ul>
</div></blockquote>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 51.697 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 51.255 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 aed264455..414d57d9a 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -504,7 +504,7 @@ for calibration. But the accuracy might be impacted.</p>
DeprecationWarning,
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 29.035 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 11.584 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 5ffeca38f..cc7ba899e 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -436,22 +436,25 @@ to your device.</p>
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
0%| | 0/132723 [00:00<?, ?KB/s]
- 5%|4 | 6010/132723 [00:00<00:02, 60095.90KB/s]
- 11%|# | 14441/132723 [00:00<00:01, 74337.36KB/s]
- 17%|#7 | 22880/132723 [00:00<00:01, 78925.54KB/s]
- 24%|##3 | 31323/132723 [00:00<00:01, 81095.09KB/s]
- 30%|### | 39842/132723 [00:00<00:01, 82569.00KB/s]
- 36%|###6 | 48277/132723 [00:00<00:01, 83171.91KB/s]
- 43%|####2 | 56678/132723 [00:00<00:00, 83443.31KB/s]
- 49%|####9 | 65163/132723 [00:00<00:00, 83888.20KB/s]
- 55%|#####5 | 73608/132723 [00:00<00:00, 84060.02KB/s]
- 62%|######1 | 82087/132723 [00:01<00:00, 84283.34KB/s]
- 68%|######8 | 90554/132723 [00:01<00:00, 84400.22KB/s]
- 75%|#######4 | 99024/132723 [00:01<00:00, 84484.57KB/s]
- 81%|######## | 107473/132723 [00:01<00:00, 84326.68KB/s]
- 87%|########7 | 115906/132723 [00:01<00:00, 84308.19KB/s]
- 94%|#########3| 124337/132723 [00:01<00:00, 84249.67KB/s]
-100%|##########| 132723/132723 [00:01<00:00, 82855.34KB/s]
+ 1%| | 940/132723 [00:00<00:14, 9328.09KB/s]
+ 2%|1 | 2610/132723 [00:00<00:09, 13593.59KB/s]
+ 4%|4 | 5746/132723 [00:00<00:05, 21630.93KB/s]
+ 9%|8 | 11442/132723 [00:00<00:03, 35483.31KB/s]
+ 15%|#4 | 19765/132723 [00:00<00:02, 52644.11KB/s]
+ 21%|##1 | 28274/132723 [00:00<00:01, 63644.58KB/s]
+ 28%|##7 | 36770/132723 [00:00<00:01, 70596.23KB/s]
+ 34%|###4 | 45212/132723 [00:00<00:01, 74989.64KB/s]
+ 41%|#### | 53778/132723 [00:00<00:01, 78320.42KB/s]
+ 47%|####6 | 62329/132723 [00:01<00:00, 80536.89KB/s]
+ 53%|#####3 | 70866/132723 [00:01<00:00, 82013.83KB/s]
+ 60%|#####9 | 79371/132723 [00:01<00:00, 82934.93KB/s]
+ 66%|######6 | 87915/132723 [00:01<00:00, 83691.48KB/s]
+ 73%|#######2 | 96517/132723 [00:01<00:00, 84390.28KB/s]
+ 79%|#######9 | 105074/132723 [00:01<00:00, 84742.93KB/s]
+ 86%|########5 | 113604/132723 [00:01<00:00, 84908.60KB/s]
+ 92%|#########2| 122134/132723 [00:01<00:00, 85023.80KB/s]
+ 98%|#########8| 130706/132723 [00:01<00:00, 85224.95KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 72578.87KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -494,7 +497,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 18.036 seconds)</p>
+<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 20.510 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 e295e5e63..05a56dfdf 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -322,7 +322,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>10:31.098</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>10:25.112</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -331,31 +331,31 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></td>
-<td><p>02:54.429</p></td>
+<td><p>03:01.189</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></td>
-<td><p>02:18.036</p></td>
+<td><p>02:20.510</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></td>
-<td><p>01:51.697</p></td>
+<td><p>01:51.255</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>01:29.035</p></td>
+<td><p>01:11.584</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>01:07.399</p></td>
+<td><p>01:08.754</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></td>
-<td><p>00:28.619</p></td>
+<td><p>00:29.501</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></td>
-<td><p>00:21.878</p></td>
+<td><p>00:22.313</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></td>
diff --git a/docs/how_to/extend_tvm/bring_your_own_datatypes.html b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
index 3940b625a..2449cddc0 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -604,7 +604,7 @@ 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.zip4b5eceb2-88b7-4508-ac97-14ab3565b274 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.zip5813f13e-1029-4f7d-a687-2c9c2dbc1d30 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>
@@ -668,7 +668,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/workspace/python/tvm/driver/build_module.py:264: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- Check failed: (lower) is false: FloatImm lowering function for target llvm type 150 not found
+ Check failed: (lower) is false: Intrinsic lowering function for target llvm, intrinsic name tir.sqrt, type 150 not found
</pre></div>
</div>
<p>When we attempt to run the model, we get a familiar error telling us that more functions need to be registered for myfloat.</p>
diff --git a/docs/how_to/extend_tvm/sg_execution_times.html b/docs/how_to/extend_tvm/sg_execution_times.html
index 70859cc05..4638bf4eb 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -322,7 +322,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-extend-tvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:39.774</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:40.139</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -331,15 +331,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></td>
-<td><p>00:36.589</p></td>
+<td><p>00:36.585</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="use_pass_instrument.html#sphx-glr-how-to-extend-tvm-use-pass-instrument-py"><span class="std std-ref">How to Use TVM Pass Instrument</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_instrument.py</span></code>)</p></td>
-<td><p>00:02.257</p></td>
+<td><p>00:02.639</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></td>
-<td><p>00:00.922</p></td>
+<td><p>00:00.908</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>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index 7fc104db6..e63afca64 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -507,10 +507,10 @@ profile the execution time of each passes.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 7271us [7271us] (46.43%; 46.43%)
-FoldScaleAxis: 8388us [6us] (53.57%; 53.57%)
- FoldConstant: 8382us [1669us] (53.53%; 99.93%)
- InferType: 6713us [6713us] (42.87%; 80.09%)
+InferType: 6703us [6703us] (45.68%; 45.68%)
+FoldScaleAxis: 7971us [6us] (54.32%; 54.32%)
+ FoldConstant: 7965us [1582us] (54.28%; 99.92%)
+ InferType: 6382us [6382us] (43.50%; 80.13%)
</pre></div>
</div>
</div>
@@ -532,10 +532,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 6723us [6723us] (44.81%; 44.81%)
-FoldScaleAxis: 8281us [5us] (55.19%; 55.19%)
- FoldConstant: 8276us [1702us] (55.16%; 99.94%)
- InferType: 6574us [6574us] (43.81%; 79.43%)
+InferType: 6341us [6341us] (44.56%; 44.56%)
+FoldScaleAxis: 7890us [5us] (55.44%; 55.44%)
+ FoldConstant: 7885us [1612us] (55.41%; 99.94%)
+ InferType: 6273us [6273us] (44.08%; 79.56%)
</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 76d59ebc2..eec692b25 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -556,7 +556,7 @@ 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: 54.142719 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 36.530970 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 c7e0bb311..a40066f49 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -898,7 +898,7 @@ 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: 6.906179 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 13.162686 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 d626e07d0..ba61fc789 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -453,8 +453,8 @@ 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: 0.019387
-Baseline: 3.341706
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019243
+Baseline: 3.344933
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -514,7 +514,7 @@ 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: 0.307661
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.301809
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -581,7 +581,7 @@ 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: 0.349376
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.344805
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -642,7 +642,7 @@ 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: 0.118716
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.120931
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -725,7 +725,7 @@ 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: 0.110635
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.111510
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -811,7 +811,7 @@ 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: 0.111683
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111780
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -901,7 +901,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt6: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">opt6_time</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.145276
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.145536
</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 b9feb5a8b..1ff453905 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -322,7 +322,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:34.598</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:34.594</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -331,15 +331,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></td>
-<td><p>00:32.358</p></td>
+<td><p>00:32.242</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>00:01.233</p></td>
+<td><p>00:01.318</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>00:01.007</p></td>
+<td><p>00:01.033</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 6d63ddf12..b1682556c 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -322,7 +322,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autoscheduler-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:22.741</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>05:12.861</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -331,27 +331,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></td>
-<td><p>02:36.299</p></td>
+<td><p>02:33.658</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>01:19.969</p></td>
+<td><p>01:20.635</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_network_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-cuda-py"><span class="std std-ref">Auto-scheduling a Neural Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_cuda.py</span></code>)</p></td>
-<td><p>00:42.898</p></td>
+<td><p>00:43.525</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>00:26.436</p></td>
+<td><p>00:17.856</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></td>
-<td><p>00:08.588</p></td>
+<td><p>00:08.662</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></td>
-<td><p>00:08.551</p></td>
+<td><p>00:08.526</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 b7adbf448..d017a124b 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
@@ -486,939 +486,659 @@ cooperative fetching, unrolling and operator fusion.</p>
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 112;
- allocate(conv2d_nchw: Pointer(local float32), float32, [7]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [672]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [7], [], scope="local", align=16)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 64;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [8]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [648]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [576]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [8], [], scope="local", align=32)[0] = 0f32
conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[2] = 0f32
conv2d_nchw_1[3] = 0f32
conv2d_nchw_1[4] = 0f32
conv2d_nchw_1[5] = 0f32
conv2d_nchw_1[6] = 0f32
- for (rc.outer.outer: int32, 0, 16) {
- for (ry.outer.outer: int32, 0, 3) {
- let cse_var_3: int32 = (rc.outer.outer*1568)
- let cse_var_2: int32 = (ry.outer.outer*7)
- let cse_var_1: int32 = (ry.outer.outer*3)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32 {
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [672], [], scope="shared")[(threadIdx.x_1*8)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*8), 21), 3) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*8), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floor [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 1)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 1), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 1), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 1), 3)*7)) + cse [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 2)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 2), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 2), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 2), 3)*7)) + cse [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 3)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 1), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod(b [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 4)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 4), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 4), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 4), 3)*7)) + cse [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 5)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 5), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 5), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 5), 3)*7)) + cse [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 6)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 2), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 2), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod(b [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 7)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 7), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 7), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 7), 3)*7)) + cse [...]
- }
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32 {
- pad_temp.shared_1[((threadIdx.x_1*8) + 256)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 256), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 256), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 256), 3)*7 [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 257)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 257), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 257), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 257), 3)*7 [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 258)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 2), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 2), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 259)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 256), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 256), 3) + 1), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 260)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 260), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 260), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 260), 3)*7 [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 261)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 3), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 3), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floormod [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 262)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 256), 3) + 2), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 256), 3) + 2), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + [...]
- pad_temp.shared_1[((threadIdx.x_1*8) + 263)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 263), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 263), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 263), 3)*7 [...]
- }
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32 {
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 512)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 512), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 512), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 512), 3) [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 513)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 3), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 3), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floorm [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 514)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 514), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 514), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 514), 3) [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 515)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 512), 3) + 1), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 512), 3) + 1), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 516)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 4), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 4), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floorm [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 517)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*8) + 517), 21), 3) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*8) + 517), 21), 3) + ry.outer.outer) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 1), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) + 517), 3) [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 518)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 512), 3) + 2), 7))) && ((ry.outer.outer + floormod((floordiv(((threadIdx.x_1*8) + 512), 3) + 2), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)))) && ((floormod(blockIdx.x, 7) + floormod(((threadIdx.x_1*2) + 2), 3)) < 8)), data[(((((cse_var_3 + (floordiv(((threadIdx.x_1*8) [...]
- }
- if @tir.likely((threadIdx.x_1 < 20), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*8) + 519)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 5), 7))) && ((ry.outer.outer + floormod((floordiv((threadIdx.x_1*8), 3) + 5), 7)) < 8)) && (1 <= (floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)))) && ((floormod(blockIdx.x, 7) + floormod((threadIdx.x_1*2), 3)) < 8)), data[(((((cse_var_3 + (floordiv((threadIdx.x_1*8), 3)*7)) + cse_var_2) + floorm [...]
- }
- }
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope="shared")[threadIdx.x_2] = kernel[(((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 32)] = kernel[(((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[(((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 96)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 4608)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 4608)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 160)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 4608)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 9216)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 9216)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 9216)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 288)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 13824)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 13824)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 352)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 13824)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 18432)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 416)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 18432)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 18432)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 480)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 23040)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 23040)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 544)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 23040)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 27648)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 608)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 27648)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 27648)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 32256)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 32256)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 736)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 32256)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 800)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 864)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 41472)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 41472)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 928)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 41472)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 46080)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 992)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 46080)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 46080)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1056)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 50688)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 50688)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 50688)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 55296)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1184)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 55296)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 55296)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1248)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 59904)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 59904)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1312)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 59904)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 64512)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1376)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 64512)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 64512)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1440)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 69120)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 69120)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1504)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 69120)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1632)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 78336)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 78336)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1696)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 78336)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 82944)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1760)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 82944)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 82944)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1824)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 87552)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 87552)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1888)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 87552)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 92160)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1952)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 92160)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 92160)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 96768)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 96768)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2080)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 96768)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 101376)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2144)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 101376)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 101376)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2208)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 105984)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 105984)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2272)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 105984)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2336)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2400)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 115200)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 115200)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2464)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 115200)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 119808)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2528)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 119808)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 119808)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2592)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 124416)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 124416)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2656)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 124416)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 129024)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2720)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 129024)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 129024)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2784)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 133632)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 133632)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2848)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 133632)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 138240)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2912)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 138240)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 138240)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 2976)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv(threadIdx.x_2, 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 142848)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 32), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3)) + 142848)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 32;
- kernel.shared_1[(threadIdx.x_2 + 3040)] = kernel[((((((floordiv(blockIdx.x, 7)*147456) + (rc.outer.outer*288)) + (floordiv((threadIdx.x_2 + 64), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3)) + 142848)]
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[0]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[9]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[12]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[18]*kernel.shared_1[(threadIdx.x*96)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*96) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*96) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*96) + 3)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*96) + 4)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*96) + 5)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*96) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*96) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*96) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[72]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[81]*kernel.shared_1[((threadIdx.x*96) + 9)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*96) + 10)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[80]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*96) + 11)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[90]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[99]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*96) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*96) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[89]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[98]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*96) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[108]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[117]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*96) + 15)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*96) + 16)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[107]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[116]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[125]*kernel.shared_1[((threadIdx.x*96) + 17)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[126]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[135]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[144]*kernel.shared_1[((threadIdx.x*96) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*96) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[134]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[143]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*96) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[153]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[162]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*96) + 21)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*96) + 22)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[152]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[161]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*96) + 23)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[171]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[180]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*96) + 24)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*96) + 25)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[170]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[179]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[188]*kernel.shared_1[((threadIdx.x*96) + 26)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[189]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[198]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[207]*kernel.shared_1[((threadIdx.x*96) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*96) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[197]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[206]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*96) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[216]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[219]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[222]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[225]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[228]*kernel.shared_1[((threadIdx.x*96) + 30)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[217]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[220]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[223]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[226]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[229]*kernel.shared_1[((threadIdx.x*96) + 31)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[215]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[218]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[221]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[224]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[227]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[230]*kernel.shared_1[((threadIdx.x*96) + 32)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[231]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[234]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[237]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[240]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[243]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[246]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[249]*kernel.shared_1[((threadIdx.x*96) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[232]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[235]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[238]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[241]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[244]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[247]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[250]*kernel.shared_1[((threadIdx.x*96) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[233]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[236]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[239]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[242]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[245]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[248]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[251]*kernel.shared_1[((threadIdx.x*96) + 35)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[252]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[255]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[258]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[261]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[264]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[267]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[270]*kernel.shared_1[((threadIdx.x*96) + 36)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[253]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[256]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[259]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[262]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[265]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[268]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[271]*kernel.shared_1[((threadIdx.x*96) + 37)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[254]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[257]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[260]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[263]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[266]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[269]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[272]*kernel.shared_1[((threadIdx.x*96) + 38)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[273]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[276]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[279]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[282]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[285]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[288]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[291]*kernel.shared_1[((threadIdx.x*96) + 39)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[274]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[277]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[280]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[283]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[286]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[289]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[292]*kernel.shared_1[((threadIdx.x*96) + 40)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[275]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[278]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[281]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[284]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[287]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[290]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[293]*kernel.shared_1[((threadIdx.x*96) + 41)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[294]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[297]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[300]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[303]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[306]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[309]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[312]*kernel.shared_1[((threadIdx.x*96) + 42)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[295]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[298]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[301]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[304]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[307]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[310]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[313]*kernel.shared_1[((threadIdx.x*96) + 43)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[296]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[299]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[302]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[305]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[308]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[311]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[314]*kernel.shared_1[((threadIdx.x*96) + 44)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[315]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[318]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[321]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[324]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[327]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[330]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[333]*kernel.shared_1[((threadIdx.x*96) + 45)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[316]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[319]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[322]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[325]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[328]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[331]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[334]*kernel.shared_1[((threadIdx.x*96) + 46)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[317]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[320]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[323]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[326]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[329]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[332]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[335]*kernel.shared_1[((threadIdx.x*96) + 47)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[336]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[339]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[342]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[345]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[348]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[351]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[354]*kernel.shared_1[((threadIdx.x*96) + 48)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[337]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[340]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[343]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[346]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[349]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[352]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[355]*kernel.shared_1[((threadIdx.x*96) + 49)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[338]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[341]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[344]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[347]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[350]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[353]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[356]*kernel.shared_1[((threadIdx.x*96) + 50)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[357]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[360]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[363]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[366]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[369]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[372]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[375]*kernel.shared_1[((threadIdx.x*96) + 51)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[358]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[361]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[364]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[367]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[370]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[373]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[376]*kernel.shared_1[((threadIdx.x*96) + 52)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[359]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[362]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[365]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[368]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[371]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[374]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[377]*kernel.shared_1[((threadIdx.x*96) + 53)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[378]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[381]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[384]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[387]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[390]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[393]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[396]*kernel.shared_1[((threadIdx.x*96) + 54)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[379]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[382]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[385]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[388]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[391]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[394]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[397]*kernel.shared_1[((threadIdx.x*96) + 55)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[380]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[383]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[386]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[389]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[392]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[395]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[398]*kernel.shared_1[((threadIdx.x*96) + 56)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[399]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[402]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[405]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[408]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[411]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[414]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[417]*kernel.shared_1[((threadIdx.x*96) + 57)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[400]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[403]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[406]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[409]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[412]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[415]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[418]*kernel.shared_1[((threadIdx.x*96) + 58)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[401]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[404]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[407]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[410]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[413]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[416]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[419]*kernel.shared_1[((threadIdx.x*96) + 59)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[420]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[423]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[426]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[429]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[432]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[435]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[438]*kernel.shared_1[((threadIdx.x*96) + 60)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[421]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[424]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[427]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[430]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[433]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[436]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[439]*kernel.shared_1[((threadIdx.x*96) + 61)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[422]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[425]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[428]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[431]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[434]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[437]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[440]*kernel.shared_1[((threadIdx.x*96) + 62)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[441]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[444]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[447]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[450]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[453]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[456]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[459]*kernel.shared_1[((threadIdx.x*96) + 63)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[442]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[445]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[448]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[451]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[454]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[457]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[460]*kernel.shared_1[((threadIdx.x*96) + 64)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[443]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[446]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[449]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[452]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[455]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[458]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[461]*kernel.shared_1[((threadIdx.x*96) + 65)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[462]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[465]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[468]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[471]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[474]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[477]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[480]*kernel.shared_1[((threadIdx.x*96) + 66)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[463]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[466]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[469]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[472]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[475]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[478]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[481]*kernel.shared_1[((threadIdx.x*96) + 67)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[464]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[467]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[470]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[473]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[476]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[479]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[482]*kernel.shared_1[((threadIdx.x*96) + 68)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[483]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[486]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[489]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[492]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[495]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[498]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[501]*kernel.shared_1[((threadIdx.x*96) + 69)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[484]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[487]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[490]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[493]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[496]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[499]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[502]*kernel.shared_1[((threadIdx.x*96) + 70)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[485]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[488]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[491]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[494]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[497]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[500]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[503]*kernel.shared_1[((threadIdx.x*96) + 71)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[504]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[507]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[510]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[513]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[516]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[519]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[522]*kernel.shared_1[((threadIdx.x*96) + 72)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[505]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[508]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[511]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[514]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[517]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[520]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[523]*kernel.shared_1[((threadIdx.x*96) + 73)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[506]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[509]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[512]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[515]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[518]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[521]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[524]*kernel.shared_1[((threadIdx.x*96) + 74)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[525]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[528]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[531]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[534]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[537]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[540]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[543]*kernel.shared_1[((threadIdx.x*96) + 75)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[526]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[529]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[532]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[535]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[538]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[541]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[544]*kernel.shared_1[((threadIdx.x*96) + 76)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[527]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[530]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[533]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[536]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[539]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[542]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[545]*kernel.shared_1[((threadIdx.x*96) + 77)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[546]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[549]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[552]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[555]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[558]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[561]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[564]*kernel.shared_1[((threadIdx.x*96) + 78)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[547]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[550]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[553]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[556]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[559]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[562]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[565]*kernel.shared_1[((threadIdx.x*96) + 79)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[548]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[551]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[554]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[557]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[560]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[563]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[566]*kernel.shared_1[((threadIdx.x*96) + 80)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[567]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[570]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[573]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[576]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[579]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[582]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[585]*kernel.shared_1[((threadIdx.x*96) + 81)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[568]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[571]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[574]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[577]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[580]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[583]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[586]*kernel.shared_1[((threadIdx.x*96) + 82)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[569]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[572]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[575]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[578]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[581]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[584]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[587]*kernel.shared_1[((threadIdx.x*96) + 83)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[588]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[591]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[594]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[597]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[600]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[603]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[606]*kernel.shared_1[((threadIdx.x*96) + 84)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[589]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[592]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[595]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[598]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[601]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[604]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[607]*kernel.shared_1[((threadIdx.x*96) + 85)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[590]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[593]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[596]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[599]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[602]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[605]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[608]*kernel.shared_1[((threadIdx.x*96) + 86)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[609]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[612]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[615]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[618]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[621]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[624]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[627]*kernel.shared_1[((threadIdx.x*96) + 87)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[610]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[613]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[616]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[619]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[622]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[625]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[628]*kernel.shared_1[((threadIdx.x*96) + 88)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[611]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[614]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[617]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[620]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[623]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[626]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[629]*kernel.shared_1[((threadIdx.x*96) + 89)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[630]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[633]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[636]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[639]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[642]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[645]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[648]*kernel.shared_1[((threadIdx.x*96) + 90)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[631]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[634]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[637]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[640]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[643]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[646]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[649]*kernel.shared_1[((threadIdx.x*96) + 91)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[632]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[635]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[638]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[641]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[644]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[647]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[650]*kernel.shared_1[((threadIdx.x*96) + 92)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[651]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[654]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[657]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[660]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[663]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[666]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[669]*kernel.shared_1[((threadIdx.x*96) + 93)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[652]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[655]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[658]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[661]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[664]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[667]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[670]*kernel.shared_1[((threadIdx.x*96) + 94)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[653]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[656]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[659]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[662]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[665]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[668]*kernel.shared_1[((threadIdx.x*96) + 95)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[671]*kernel.shared_1[((threadIdx.x*96) + 95)]))
+ conv2d_nchw_1[7] = 0f32
+ for (rc.outer.outer: int32, 0, 64) {
+ let cse_var_2: int32 = (rc.outer.outer*392)
+ let cse_var_1: int32 = (rc.outer.outer*72)
+ {
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [648], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else((((9 <= threadIdx.x_1) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[(((cse_var_2 + (floordiv(threadIdx.x_1, 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 49), 81)) && (floormod((threadIdx.x_1 + 49), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 49), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 49), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 8), 9)) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 98), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 98), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 147), 81)) && (floormod((threadIdx.x_1 + 66), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 3), 9))) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 147), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 147), 81), 9)*7)) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 196), 81)) && (floormod((threadIdx.x_1 + 34), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 196), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 196), 81), 9)*7)) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 245)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 245), 81)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 245), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 245), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 294), 81)) && (floormod((threadIdx.x_1 + 51), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 294), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 294), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 343)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 1), 9)) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 343), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 343), 81), 9)*7)) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 392), 81)) && (floormod((threadIdx.x_1 + 68), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 392), 81), 9)*7)) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 441)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 9) + 4), 9)) && (floormod((threadIdx.x_1 + 36), 81) < 72)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 441), 81)*49)) + (floormod((floordiv(threadIdx.x_1, 9) + 4), 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 490), 81)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 490), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 490), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 539)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 539), 81)) && (floormod((threadIdx.x_1 + 53), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 539), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 539), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ pad_temp.shared_1[(threadIdx.x_1 + 588)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 3), 9)) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 588), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 588), 81), 9)*7)) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ if @tir.likely((threadIdx.x_1 < 11), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 637)] = @tir.if_then_else((((floormod((threadIdx.x_1 + 70), 81) < 72) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 637), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 637), 81), 9)*7)) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
}
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1: Buffer(kernel.shared, float32, [576], [], scope="shared")[threadIdx.x_2] = kernel[(((blockIdx.x*36864) + cse_var_1) + threadIdx.x_2)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 49), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 49), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 98), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 98), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 147)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 147), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 1), 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 196)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 196), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 196), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 245)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 245), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 245), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 294)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 294), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 2), 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 343)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 343), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 343), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 392), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 392), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 441)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 441), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 3), 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ kernel.shared_1[(threadIdx.x_2 + 490)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 490), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 490), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+ if @tir.likely((threadIdx.x_2 < 37), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 539)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 539), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 539), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ }
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[0]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[72]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[144]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[216]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[288]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[360]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[432]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7))]*kernel.shared_1[504]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[1]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[73]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[145]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[217]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[289]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[361]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[433]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[505]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[2]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[74]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[146]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[218]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[290]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[362]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[434]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[506]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[9]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[81]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[153]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[225]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[297]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[369]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[441]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 81)]*kernel.shared_1[513]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[10]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[82]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[154]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[226]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[298]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[370]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[442]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 82)]*kernel.shared_1[514]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[11]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[83]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[155]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[227]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[299]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[371]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[443]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 83)]*kernel.shared_1[515]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[18]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[90]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[162]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[234]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[306]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[378]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[450]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 162)]*kernel.shared_1[522]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[19]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[91]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[163]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[235]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[307]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[379]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[451]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 163)]*kernel.shared_1[523]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[20]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[92]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[164]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[236]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[308]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[380]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[452]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 164)]*kernel.shared_1[524]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[27]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[99]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[171]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[243]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[315]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[387]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[459]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 243)]*kernel.shared_1[531]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[28]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[100]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[172]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[244]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[316]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[388]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[460]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 244)]*kernel.shared_1[532]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[29]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[101]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[173]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[245]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[317]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[389]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[461]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 245)]*kernel.shared_1[533]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[3]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[75]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[147]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[219]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[291]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[363]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[435]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[507]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[4]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[76]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[148]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[220]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[292]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[364]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[436]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[508]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[5]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[77]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[149]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[221]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[293]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[365]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[437]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[509]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[12]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[84]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[156]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[228]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[300]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[372]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[444]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 90)]*kernel.shared_1[516]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[13]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[85]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[157]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[229]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[301]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[373]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[445]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[517]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[14]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[86]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[158]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[230]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[302]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[374]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[446]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 92)]*kernel.shared_1[518]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[21]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[93]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[165]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[237]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[309]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[381]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[453]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 171)]*kernel.shared_1[525]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[22]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[94]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[166]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[238]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[310]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[382]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[454]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 172)]*kernel.shared_1[526]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[23]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[95]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[167]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[239]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[311]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[383]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[455]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 173)]*kernel.shared_1[527]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[30]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[102]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[174]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[246]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[318]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[390]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[462]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 252)]*kernel.shared_1[534]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[31]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[103]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[175]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[247]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[319]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[391]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[463]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 253)]*kernel.shared_1[535]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[32]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[104]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[176]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[248]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[320]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[392]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[464]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 254)]*kernel.shared_1[536]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[6]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[78]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[150]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[222]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[294]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[366]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[438]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[510]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[7]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[79]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[151]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[223]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[295]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[367]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[439]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[511]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[8]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[80]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[152]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[224]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[296]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[368]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[440]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[512]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[15]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[87]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[159]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[231]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[303]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[375]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[447]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 99)]*kernel.shared_1[519]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[16]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[88]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[160]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[232]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[304]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[376]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[448]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 100)]*kernel.shared_1[520]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[17]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[89]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[161]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[233]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[305]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[377]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[449]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 101)]*kernel.shared_1[521]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[24]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[96]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[168]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[240]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[312]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[384]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[456]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 180)]*kernel.shared_1[528]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[25]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[97]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[169]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[241]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[313]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[385]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[457]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 181)]*kernel.shared_1[529]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[26]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[98]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[170]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[242]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[314]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[386]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[458]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 182)]*kernel.shared_1[530]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[33]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[105]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[177]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[249]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[321]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[393]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[465]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 261)]*kernel.shared_1[537]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[34]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[106]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[178]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[250]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[322]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[394]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[466]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 262)]*kernel.shared_1[538]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[35]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[107]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[179]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[251]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[323]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[395]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[467]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 263)]*kernel.shared_1[539]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[36]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[108]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[180]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[252]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[324]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[396]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[468]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 324)]*kernel.shared_1[540]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[37]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[109]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[181]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[253]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[325]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[397]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[469]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 325)]*kernel.shared_1[541]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[38]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[110]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[182]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[254]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[326]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[398]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[470]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 326)]*kernel.shared_1[542]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[45]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[117]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[189]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[261]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[333]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[405]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[477]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 405)]*kernel.shared_1[549]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[46]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[118]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[190]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[262]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[334]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[406]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[478]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 406)]*kernel.shared_1[550]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[47]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[119]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[191]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[263]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[335]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[407]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[479]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 407)]*kernel.shared_1[551]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[54]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[126]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[198]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[270]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[342]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[414]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[486]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 486)]*kernel.shared_1[558]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[55]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[127]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[199]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[271]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[343]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[415]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[487]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 487)]*kernel.shared_1[559]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[56]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[128]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[200]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[272]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[344]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[416]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[488]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 488)]*kernel.shared_1[560]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[63]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[135]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[207]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[279]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[351]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[423]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[495]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 567)]*kernel.shared_1[567]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[64]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[136]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[208]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[280]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[352]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[424]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[496]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 568)]*kernel.shared_1[568]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[65]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[137]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[209]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[281]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[353]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[425]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[497]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 569)]*kernel.shared_1[569]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[39]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[111]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[183]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[255]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[327]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[399]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[471]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 333)]*kernel.shared_1[543]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[40]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[112]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[184]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[256]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[328]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[400]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[472]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 334)]*kernel.shared_1[544]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[41]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[113]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[185]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[257]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[329]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[401]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[473]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 335)]*kernel.shared_1[545]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[48]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[120]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[192]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[264]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[336]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[408]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[480]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 414)]*kernel.shared_1[552]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[49]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[121]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[193]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[265]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[337]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[409]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[481]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 415)]*kernel.shared_1[553]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[50]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[122]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[194]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[266]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[338]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[410]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[482]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 416)]*kernel.shared_1[554]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[57]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[129]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[201]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[273]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[345]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[417]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[489]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 495)]*kernel.shared_1[561]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[58]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[130]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[202]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[274]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[346]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[418]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[490]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 496)]*kernel.shared_1[562]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[59]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[131]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[203]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[275]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[347]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[419]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[491]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 497)]*kernel.shared_1[563]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[66]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[138]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[210]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[282]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[354]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[426]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[498]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 576)]*kernel.shared_1[570]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[67]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[139]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[211]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[283]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[355]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[427]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[499]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 577)]*kernel.shared_1[571]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[68]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[140]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[212]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[284]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[356]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[428]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[500]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 578)]*kernel.shared_1[572]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[42]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[114]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[186]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[258]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[330]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[402]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[474]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 342)]*kernel.shared_1[546]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[43]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[115]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[187]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[259]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[331]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[403]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[475]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 343)]*kernel.shared_1[547]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[44]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[116]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[188]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[260]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[332]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[404]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[476]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 344)]*kernel.shared_1[548]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[51]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[123]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[195]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[267]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[339]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[411]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[483]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 423)]*kernel.shared_1[555]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[52]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[124]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[196]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[268]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[340]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[412]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[484]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 424)]*kernel.shared_1[556]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[53]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[125]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[197]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[269]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[341]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[413]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[485]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 425)]*kernel.shared_1[557]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[60]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[132]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[204]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[276]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[348]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[420]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[492]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 504)]*kernel.shared_1[564]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[61]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[133]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[205]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[277]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[349]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[421]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[493]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 505)]*kernel.shared_1[565]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[62]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[134]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[206]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[278]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[350]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[422]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[494]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 506)]*kernel.shared_1[566]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[69]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[141]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[213]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[285]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[357]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[429]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[501]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 585)]*kernel.shared_1[573]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[70]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[142]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[214]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[286]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[358]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[430]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[502]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 586)]*kernel.shared_1[574]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[71]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[143]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[215]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[287]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[359]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[431]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[503]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((floordiv(threadIdx.x, 7)*9) + floormod(threadIdx.x, 7)) + 587)]*kernel.shared_1[575]))
}
}
- for (i2.inner: int32, 0, 7) {
- compute[((((floordiv(blockIdx.x, 7)*1568) + (threadIdx.x*49)) + (i2.inner*7)) + floormod(blockIdx.x, 7))] = max((conv2d_nchw_1[i2.inner] + bias[((floordiv(blockIdx.x, 7)*32) + threadIdx.x)]), 0f32)
+ for (i1.inner: int32, 0, 8) {
+ compute[(((blockIdx.x*392) + (i1.inner*49)) + threadIdx.x)] = max((conv2d_nchw_1[i1.inner] + bias[((blockIdx.x*8) + i1.inner)]), 0f32)
}
}
}
@@ -1455,7 +1175,7 @@ cooperative fetching, unrolling and operator fusion.</p>
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.404 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.220 ms
</pre></div>
</div>
</div>
@@ -1484,36 +1204,36 @@ conv2d_nchw_nn_o_i, conv2d_nchw_nn_i = s[conv2d_nchw].split(conv2d_nchw_nn, fact
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
-conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
+conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=8)
conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
-conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=32)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=1)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
-conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=7)
+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=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=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=16)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
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=3)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=32)
+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=1)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
-compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=7)
-compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
+compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
-compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=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)
@@ -1533,12 +1253,12 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=32)
+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=49)
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=8)
+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=32)
+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=49)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 1024)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
@@ -1558,10 +1278,10 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(32) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[7];
- __shared__ float pad_temp_shared[672];
- __shared__ float kernel_shared[3072];
+extern "C" __global__ void __launch_bounds__(49) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[8];
+ __shared__ float pad_temp_shared[648];
+ __shared__ float kernel_shared[576];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
@@ -1569,822 +1289,619 @@ extern "C" __global__ void __launch_bounds__(32) default_function_kern
conv2d_nchw[4] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
- for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
- __syncthreads();
- pad_temp_shared[(((int)threadIdx.x) * 8)] = (((((1 <= ((((((int)threadIdx.x) * 8) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) * 8) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((( [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 1)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 1) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 1) / 3) * 7)) + (ry_outer_outer * [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 2)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 2) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 2) / 3) * 7)) + (ry_outer_outer * [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 3)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 1) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockId [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 4)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 4) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 4) / 3) * 7)) + (ry_outer_outer * [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 5)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 5) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 5) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 5) / 3) * 7)) + (ry_outer_outer * [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 6)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 2) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 2) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockId [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 7)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 7) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 7) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 7) / 3) * 7)) + (ry_outer_outer * [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 256)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 4) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 256) / 3) * 7)) + (ry_outer_oute [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 257)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 5) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 5) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 257) / 3) * 7)) + (ry_outer_oute [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 258)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 2) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 2) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)block [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 259)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 8) + 256) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 8) + 256) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 256) / 3) * 7)) + [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 260)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 8) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 8) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 260) / 3) * 7)) + (ry_outer_oute [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 261)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 3) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 3) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)block [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 262)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 8) + 256) / 3) + 2) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 8) + 256) / 3) + 2) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 256) / 3) * 7)) + [...]
- pad_temp_shared[((((int)threadIdx.x) * 8) + 263)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 11) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 11) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 263) / 3) * 7)) + (ry_outer_ou [...]
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 512)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 8) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 8) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 512) / 3) * 7)) + (ry_outer_ou [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 513)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 3) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 3) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blo [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 514)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 10) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 10) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 514) / 3) * 7)) + (ry_outer_ [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 515)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 8) + 512) / 3) + 1) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 8) + 512) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 512) / 3) * 7)) [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 516)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 4) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 4) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blo [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 517)] = (((((1 <= (((((((int)threadIdx.x) * 8) + 13) % 21) / 3) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 8) + 13) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 517) / 3) * 7)) + (ry_outer_ [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 518)] = (((((1 <= (ry_outer_outer + (((((((int)threadIdx.x) * 8) + 512) / 3) + 2) % 7))) && ((ry_outer_outer + (((((((int)threadIdx.x) * 8) + 512) / 3) + 2) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + (((((int)threadIdx.x) * 2) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((((int)threadIdx.x) * 8) + 512) / 3) * 7)) [...]
- }
- if (((int)threadIdx.x) < 20) {
- pad_temp_shared[((((int)threadIdx.x) * 8) + 519)] = (((((1 <= (ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 5) % 7))) && ((ry_outer_outer + ((((((int)threadIdx.x) * 8) / 3) + 5) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) * 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) * 8) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blo [...]
- }
- kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
- kernel_shared[(((int)threadIdx.x) + 32)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 64)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 96)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 4608)];
- kernel_shared[(((int)threadIdx.x) + 128)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 4608)];
- kernel_shared[(((int)threadIdx.x) + 160)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 4608)];
- kernel_shared[(((int)threadIdx.x) + 192)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 9216)];
- kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 9216)];
- kernel_shared[(((int)threadIdx.x) + 256)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 9216)];
- kernel_shared[(((int)threadIdx.x) + 288)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 13824)];
- kernel_shared[(((int)threadIdx.x) + 320)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 13824)];
- kernel_shared[(((int)threadIdx.x) + 352)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 13824)];
- kernel_shared[(((int)threadIdx.x) + 384)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 18432)];
- kernel_shared[(((int)threadIdx.x) + 416)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 18432)];
- kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 18432)];
- kernel_shared[(((int)threadIdx.x) + 480)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 23040)];
- kernel_shared[(((int)threadIdx.x) + 512)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 23040)];
- kernel_shared[(((int)threadIdx.x) + 544)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 23040)];
- kernel_shared[(((int)threadIdx.x) + 576)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 27648)];
- kernel_shared[(((int)threadIdx.x) + 608)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 27648)];
- kernel_shared[(((int)threadIdx.x) + 640)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 27648)];
- kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 32256)];
- kernel_shared[(((int)threadIdx.x) + 704)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 32256)];
- kernel_shared[(((int)threadIdx.x) + 736)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 32256)];
- kernel_shared[(((int)threadIdx.x) + 768)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((int)threadIdx.x) + 800)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 36864)];
- kernel_shared[(((int)threadIdx.x) + 832)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 36864)];
- kernel_shared[(((int)threadIdx.x) + 864)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 41472)];
- kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 41472)];
- kernel_shared[(((int)threadIdx.x) + 928)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 41472)];
- kernel_shared[(((int)threadIdx.x) + 960)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 46080)];
- kernel_shared[(((int)threadIdx.x) + 992)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 46080)];
- kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 46080)];
- kernel_shared[(((int)threadIdx.x) + 1056)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 50688)];
- kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 50688)];
- kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 50688)];
- kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 55296)];
- kernel_shared[(((int)threadIdx.x) + 1184)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 55296)];
- kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 55296)];
- kernel_shared[(((int)threadIdx.x) + 1248)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 59904)];
- kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 59904)];
- kernel_shared[(((int)threadIdx.x) + 1312)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 59904)];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 64512)];
- kernel_shared[(((int)threadIdx.x) + 1376)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 64512)];
- kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 64512)];
- kernel_shared[(((int)threadIdx.x) + 1440)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 69120)];
- kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 69120)];
- kernel_shared[(((int)threadIdx.x) + 1504)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 69120)];
- kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 73728)];
- kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 73728)];
- kernel_shared[(((int)threadIdx.x) + 1632)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 78336)];
- kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 78336)];
- kernel_shared[(((int)threadIdx.x) + 1696)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 78336)];
- kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 82944)];
- kernel_shared[(((int)threadIdx.x) + 1760)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 82944)];
- kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 82944)];
- kernel_shared[(((int)threadIdx.x) + 1824)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 87552)];
- kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 87552)];
- kernel_shared[(((int)threadIdx.x) + 1888)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 87552)];
- kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 92160)];
- kernel_shared[(((int)threadIdx.x) + 1952)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 92160)];
- kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 92160)];
- kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 96768)];
- kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 96768)];
- kernel_shared[(((int)threadIdx.x) + 2080)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 96768)];
- kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 101376)];
- kernel_shared[(((int)threadIdx.x) + 2144)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 101376)];
- kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 101376)];
- kernel_shared[(((int)threadIdx.x) + 2208)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 105984)];
- kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 105984)];
- kernel_shared[(((int)threadIdx.x) + 2272)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 105984)];
- kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((int)threadIdx.x) + 2336)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 110592)];
- kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 110592)];
- kernel_shared[(((int)threadIdx.x) + 2400)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 115200)];
- kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 115200)];
- kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 115200)];
- kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 119808)];
- kernel_shared[(((int)threadIdx.x) + 2528)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 119808)];
- kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 119808)];
- kernel_shared[(((int)threadIdx.x) + 2592)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 124416)];
- kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 124416)];
- kernel_shared[(((int)threadIdx.x) + 2656)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 124416)];
- kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 129024)];
- kernel_shared[(((int)threadIdx.x) + 2720)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 129024)];
- kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 129024)];
- kernel_shared[(((int)threadIdx.x) + 2784)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 133632)];
- kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 133632)];
- kernel_shared[(((int)threadIdx.x) + 2848)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 133632)];
- kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 138240)];
- kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 138240)];
- kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 138240)];
- kernel_shared[(((int)threadIdx.x) + 2976)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 142848)];
- kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3)) + 142848)];
- kernel_shared[(((int)threadIdx.x) + 3040)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 64) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3)) + 142848)];
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[9] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[12] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[18] * kernel_shared[(((int)threadIdx.x) * 96)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 96) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 96) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 96) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 96) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 96) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 96) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 96) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 96) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[72] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[81] * kernel_shared[((((int)threadIdx.x) * 96) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 96) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[80] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 96) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[90] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[99] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 96) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 96) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[89] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[98] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 96) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[108] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[117] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 96) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 96) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[107] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[116] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[125] * kernel_shared[((((int)threadIdx.x) * 96) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[126] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[135] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[144] * kernel_shared[((((int)threadIdx.x) * 96) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 96) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[134] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[143] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 96) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[153] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[162] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 96) + 21)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 96) + 22)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[152] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[161] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 96) + 23)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[171] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[180] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 96) + 24)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 96) + 25)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[170] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[179] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[188] * kernel_shared[((((int)threadIdx.x) * 96) + 26)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[189] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[198] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[207] * kernel_shared[((((int)threadIdx.x) * 96) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 96) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[197] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[206] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 96) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[216] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[219] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[222] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[225] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[228] * kernel_shared[((((int)threadIdx.x) * 96) + 30)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[217] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[220] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[223] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[226] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[229] * kernel_shared[((((int)threadIdx.x) * 96) + 31)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[215] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[218] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[221] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[224] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[227] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[230] * kernel_shared[((((int)threadIdx.x) * 96) + 32)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[231] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[234] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[237] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[240] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[243] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[246] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[249] * kernel_shared[((((int)threadIdx.x) * 96) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[232] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[235] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[238] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[241] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[244] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[247] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[250] * kernel_shared[((((int)threadIdx.x) * 96) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[233] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[236] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[239] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[242] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[245] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[248] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[251] * kernel_shared[((((int)threadIdx.x) * 96) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[252] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[255] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[258] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[261] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[264] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[267] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[270] * kernel_shared[((((int)threadIdx.x) * 96) + 36)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[253] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[256] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[259] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[262] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[265] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[268] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[271] * kernel_shared[((((int)threadIdx.x) * 96) + 37)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[254] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[257] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[260] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[263] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[266] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[269] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[272] * kernel_shared[((((int)threadIdx.x) * 96) + 38)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[273] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[276] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[279] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[282] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[285] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[288] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[291] * kernel_shared[((((int)threadIdx.x) * 96) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[274] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[277] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[280] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[283] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[286] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[289] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[292] * kernel_shared[((((int)threadIdx.x) * 96) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[275] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[278] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[281] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[284] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[287] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[290] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[293] * kernel_shared[((((int)threadIdx.x) * 96) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[294] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[297] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[300] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[303] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[306] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[309] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[312] * kernel_shared[((((int)threadIdx.x) * 96) + 42)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[295] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[298] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[301] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[304] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[307] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[310] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[313] * kernel_shared[((((int)threadIdx.x) * 96) + 43)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[296] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[299] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[302] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[305] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[308] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[311] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[314] * kernel_shared[((((int)threadIdx.x) * 96) + 44)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[315] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[318] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[321] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[324] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[327] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[330] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[333] * kernel_shared[((((int)threadIdx.x) * 96) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[316] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[319] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[322] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[325] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[328] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[331] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[334] * kernel_shared[((((int)threadIdx.x) * 96) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[317] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[320] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[323] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[326] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[329] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[332] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[335] * kernel_shared[((((int)threadIdx.x) * 96) + 47)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[336] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[339] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[342] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[345] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[348] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[351] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[354] * kernel_shared[((((int)threadIdx.x) * 96) + 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[337] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[340] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[343] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[346] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[349] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[352] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[355] * kernel_shared[((((int)threadIdx.x) * 96) + 49)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[338] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[341] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[344] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[347] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[350] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[353] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[356] * kernel_shared[((((int)threadIdx.x) * 96) + 50)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[357] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[360] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[363] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[366] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[369] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[372] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[375] * kernel_shared[((((int)threadIdx.x) * 96) + 51)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[358] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[361] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[364] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[367] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[370] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[373] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[376] * kernel_shared[((((int)threadIdx.x) * 96) + 52)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[359] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[362] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[365] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[368] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[371] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[374] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[377] * kernel_shared[((((int)threadIdx.x) * 96) + 53)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[378] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[381] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[384] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[387] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[390] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[393] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[396] * kernel_shared[((((int)threadIdx.x) * 96) + 54)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[379] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[382] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[385] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[388] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[391] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[394] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[397] * kernel_shared[((((int)threadIdx.x) * 96) + 55)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[380] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[383] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[386] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[389] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[392] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[395] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[398] * kernel_shared[((((int)threadIdx.x) * 96) + 56)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[399] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[402] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[405] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[408] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[411] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[414] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[417] * kernel_shared[((((int)threadIdx.x) * 96) + 57)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[400] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[403] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[406] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[409] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[412] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[415] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[418] * kernel_shared[((((int)threadIdx.x) * 96) + 58)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[401] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[404] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[407] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[410] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[413] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[416] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[419] * kernel_shared[((((int)threadIdx.x) * 96) + 59)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[420] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[423] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[426] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[429] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[432] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[435] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[438] * kernel_shared[((((int)threadIdx.x) * 96) + 60)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[421] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[424] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[427] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[430] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[433] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[436] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[439] * kernel_shared[((((int)threadIdx.x) * 96) + 61)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[422] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[425] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[428] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[431] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[434] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[437] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[440] * kernel_shared[((((int)threadIdx.x) * 96) + 62)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[441] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[444] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[447] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[450] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[453] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[456] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[459] * kernel_shared[((((int)threadIdx.x) * 96) + 63)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[442] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[445] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[448] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[451] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[454] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[457] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[460] * kernel_shared[((((int)threadIdx.x) * 96) + 64)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[443] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[446] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[449] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[452] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[455] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[458] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[461] * kernel_shared[((((int)threadIdx.x) * 96) + 65)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[462] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[465] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[468] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[471] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[474] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[477] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[480] * kernel_shared[((((int)threadIdx.x) * 96) + 66)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[463] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[466] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[469] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[472] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[475] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[478] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[481] * kernel_shared[((((int)threadIdx.x) * 96) + 67)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[464] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[467] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[470] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[473] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[476] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[479] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[482] * kernel_shared[((((int)threadIdx.x) * 96) + 68)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[483] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[486] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[489] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[492] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[495] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[498] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[501] * kernel_shared[((((int)threadIdx.x) * 96) + 69)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[484] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[487] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[490] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[493] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[496] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[499] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[502] * kernel_shared[((((int)threadIdx.x) * 96) + 70)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[485] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[488] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[491] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[494] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[497] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[500] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[503] * kernel_shared[((((int)threadIdx.x) * 96) + 71)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[504] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[507] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[510] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[513] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[516] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[519] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[522] * kernel_shared[((((int)threadIdx.x) * 96) + 72)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[505] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[508] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[511] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[514] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[517] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[520] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[523] * kernel_shared[((((int)threadIdx.x) * 96) + 73)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[506] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[509] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[512] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[515] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[518] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[521] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[524] * kernel_shared[((((int)threadIdx.x) * 96) + 74)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[525] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[528] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[531] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[534] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[537] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[540] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[543] * kernel_shared[((((int)threadIdx.x) * 96) + 75)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[526] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[529] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[532] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[535] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[538] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[541] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[544] * kernel_shared[((((int)threadIdx.x) * 96) + 76)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[527] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[530] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[533] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[536] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[539] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[542] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[545] * kernel_shared[((((int)threadIdx.x) * 96) + 77)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[546] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[549] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[552] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[555] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[558] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[561] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[564] * kernel_shared[((((int)threadIdx.x) * 96) + 78)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[547] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[550] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[553] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[556] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[559] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[562] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[565] * kernel_shared[((((int)threadIdx.x) * 96) + 79)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[548] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[551] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[554] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[557] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[560] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[563] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[566] * kernel_shared[((((int)threadIdx.x) * 96) + 80)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[567] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[570] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[573] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[576] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[579] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[582] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[585] * kernel_shared[((((int)threadIdx.x) * 96) + 81)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[568] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[571] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[574] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[577] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[580] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[583] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[586] * kernel_shared[((((int)threadIdx.x) * 96) + 82)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[569] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[572] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[575] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[578] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[581] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[584] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[587] * kernel_shared[((((int)threadIdx.x) * 96) + 83)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[588] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[591] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[594] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[597] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[600] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[603] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[606] * kernel_shared[((((int)threadIdx.x) * 96) + 84)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[589] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[592] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[595] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[598] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[601] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[604] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[607] * kernel_shared[((((int)threadIdx.x) * 96) + 85)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[590] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[593] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[596] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[599] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[602] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[605] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[608] * kernel_shared[((((int)threadIdx.x) * 96) + 86)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[609] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[612] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[615] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[618] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[621] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[624] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[627] * kernel_shared[((((int)threadIdx.x) * 96) + 87)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[610] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[613] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[616] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[619] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[622] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[625] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[628] * kernel_shared[((((int)threadIdx.x) * 96) + 88)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[611] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[614] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[617] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[620] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[623] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[626] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[629] * kernel_shared[((((int)threadIdx.x) * 96) + 89)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[630] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[633] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[636] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[639] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[642] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[645] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[648] * kernel_shared[((((int)threadIdx.x) * 96) + 90)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[631] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[634] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[637] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[640] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[643] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[646] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[649] * kernel_shared[((((int)threadIdx.x) * 96) + 91)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[632] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[635] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[638] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[641] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[644] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[647] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[650] * kernel_shared[((((int)threadIdx.x) * 96) + 92)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[651] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[654] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[657] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[660] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[663] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[666] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[669] * kernel_shared[((((int)threadIdx.x) * 96) + 93)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[652] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[655] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[658] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[661] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[664] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[667] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[670] * kernel_shared[((((int)threadIdx.x) * 96) + 94)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[653] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[656] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[659] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[662] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[665] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[668] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[671] * kernel_shared[((((int)threadIdx.x) * 96) + 95)]));
+ conv2d_nchw[7] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[((int)threadIdx.x)] = ((((9 <= ((int)threadIdx.x)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 49)] = (((((9 <= ((((int)threadIdx.x) + 49) % 81)) && (((((int)threadIdx.x) + 49) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 49) / 81) * 49)) + ((((((int)threadIdx.x) + 49) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 98)] = (((1 <= ((((int)threadIdx.x) + 8) % 9)) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((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) + 147)] = (((((9 <= ((((int)threadIdx.x) + 66) % 81)) && (((((int)threadIdx.x) + 66) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 147) / 81) * 49)) + ((((((int)threadIdx.x) + 66) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 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 * 392) + (((((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) + 245)] = ((((9 <= ((((int)threadIdx.x) + 2) % 81)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 245) / 81) * 49)) + ((((((int)threadIdx.x) + 2) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 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 * 392) + (((((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) + 343)] = (((1 <= ((((int)threadIdx.x) + 1) % 9)) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 343) / 81) * 49)) + ((((((int)threadIdx.x) + 19) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 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 * 392) + (((((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) + 441)] = (((((1 <= (((((int)threadIdx.x) / 9) + 4) % 9)) && (((((int)threadIdx.x) + 36) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 441) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 4) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 490)] = ((((9 <= ((((int)threadIdx.x) + 4) % 81)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((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) + 539)] = (((((9 <= ((((int)threadIdx.x) + 53) % 81)) && (((((int)threadIdx.x) + 53) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 539) / 81) * 49)) + ((((((int)threadIdx.x) + 53) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 588)] = (((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 588) / 81) * 49)) + ((((((int)threadIdx.x) + 21) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 11) {
+ pad_temp_shared[(((int)threadIdx.x) + 637)] = ((((((int)threadIdx.x) < 2) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 637) / 81) * 49)) + ((((((int)threadIdx.x) + 70) % 81) / 9) * 7)) + (((int)threadIdx.x) + 7)) - 8)] : 0.000000e+00f);
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 36864) + (rc_outer_outer * 72)) + ((int)threadIdx.x))];
+ kernel_shared[(((int)threadIdx.x) + 49)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 49) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 49) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 98)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 98) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 26) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 147)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 147) / 72) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) / 3) + 1) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 196) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 52) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 245)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 245) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 29) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 294)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 294) / 72) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) / 3) + 2) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 343)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 343) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 55) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 392) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 32) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 441)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 441) / 72) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) / 3) + 3) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 490)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 490) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 58) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ if (((int)threadIdx.x) < 37) {
+ kernel_shared[(((int)threadIdx.x) + 539)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 539) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 35) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
}
+ __syncthreads();
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[0]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[72]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[144]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[216]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[288]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[360]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[432]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7))] * kernel_shared[504]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[1]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[73]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[145]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[217]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[289]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[361]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[433]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[505]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[2]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[74]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[146]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[218]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[290]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[362]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[434]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[506]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[9]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[81]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[153]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[225]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[297]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[369]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[441]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[513]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[10]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[82]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[154]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[226]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[298]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[370]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[442]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[514]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[11]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[83]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[155]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[227]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[299]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[371]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[443]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[515]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[18]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[90]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[162]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[234]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[306]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[378]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[450]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 162)] * kernel_shared[522]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[19]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[91]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[163]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[235]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[307]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[379]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[451]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 163)] * kernel_shared[523]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[20]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[92]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[164]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[236]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[308]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[380]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[452]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 164)] * kernel_shared[524]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[27]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[99]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[171]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[243]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[315]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[387]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[459]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 243)] * kernel_shared[531]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[28]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[100]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[172]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[244]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[316]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[388]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[460]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 244)] * kernel_shared[532]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[29]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[101]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[173]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[245]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[317]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[389]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[461]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 245)] * kernel_shared[533]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[3]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[75]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[147]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[219]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[291]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[363]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[435]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[507]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[4]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[76]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[148]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[220]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[292]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[364]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[436]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[508]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[5]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[77]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[149]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[221]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[293]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[365]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[437]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[509]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[12]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[84]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[156]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[228]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[300]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[372]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[444]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 90)] * kernel_shared[516]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[13]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[85]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[157]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[229]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[301]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[373]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[445]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[517]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[14]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[86]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[158]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[230]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[302]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[374]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[446]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 92)] * kernel_shared[518]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[21]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[93]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[165]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[237]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[309]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[381]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[453]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 171)] * kernel_shared[525]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[22]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[94]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[166]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[238]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[310]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[382]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[454]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 172)] * kernel_shared[526]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[23]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[95]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[167]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[239]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[311]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[383]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[455]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 173)] * kernel_shared[527]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[30]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[102]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[174]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[246]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[318]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[390]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[462]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 252)] * kernel_shared[534]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[31]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[103]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[175]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[247]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[319]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[391]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[463]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 253)] * kernel_shared[535]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[32]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[104]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[176]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[248]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[320]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[392]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[464]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 254)] * kernel_shared[536]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[6]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[78]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[150]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[222]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[294]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[366]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[438]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[510]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[7]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[79]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[151]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[223]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[295]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[367]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[439]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[511]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[8]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[80]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[152]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[224]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[296]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[368]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[440]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[512]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[15]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[87]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[159]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[231]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[303]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[375]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[447]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 99)] * kernel_shared[519]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[16]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[88]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[160]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[232]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[304]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[376]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[448]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 100)] * kernel_shared[520]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[17]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[89]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[161]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[233]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[305]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[377]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[449]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 101)] * kernel_shared[521]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[24]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[96]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[168]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[240]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[312]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[384]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[456]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 180)] * kernel_shared[528]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[25]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[97]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[169]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[241]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[313]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[385]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[457]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 181)] * kernel_shared[529]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[26]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[98]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[170]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[242]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[314]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[386]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[458]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 182)] * kernel_shared[530]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[33]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[105]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[177]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[249]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[321]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[393]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[465]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 261)] * kernel_shared[537]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[34]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[106]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[178]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[250]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[322]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[394]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[466]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 262)] * kernel_shared[538]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[35]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[107]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[179]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[251]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[323]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[395]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[467]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 263)] * kernel_shared[539]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[36]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[108]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[180]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[252]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[324]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[396]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[468]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 324)] * kernel_shared[540]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[37]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[109]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[181]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[253]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[325]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[397]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[469]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 325)] * kernel_shared[541]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[38]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[110]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[182]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[254]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[326]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[398]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[470]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 326)] * kernel_shared[542]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[45]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[117]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[189]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[261]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[333]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[405]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[477]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 405)] * kernel_shared[549]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[46]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[118]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[190]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[262]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[334]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[406]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[478]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 406)] * kernel_shared[550]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[47]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[119]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[191]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[263]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[335]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[407]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[479]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 407)] * kernel_shared[551]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[54]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[126]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[198]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[270]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[342]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[414]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[486]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 486)] * kernel_shared[558]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[55]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[127]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[199]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[271]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[343]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[415]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[487]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 487)] * kernel_shared[559]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[56]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[128]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[200]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[272]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[344]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[416]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[488]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 488)] * kernel_shared[560]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[63]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[135]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[207]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[279]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[351]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[423]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[495]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 567)] * kernel_shared[567]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[64]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[136]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[208]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[280]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[352]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[424]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[496]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 568)] * kernel_shared[568]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[65]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[137]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[209]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[281]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[353]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[425]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[497]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 569)] * kernel_shared[569]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[39]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[111]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[183]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[255]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[327]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[399]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[471]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 333)] * kernel_shared[543]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[40]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[112]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[184]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[256]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[328]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[400]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[472]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 334)] * kernel_shared[544]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[41]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[113]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[185]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[257]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[329]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[401]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[473]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 335)] * kernel_shared[545]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[48]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[120]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[192]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[264]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[336]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[408]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[480]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 414)] * kernel_shared[552]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[49]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[121]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[193]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[265]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[337]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[409]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[481]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 415)] * kernel_shared[553]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[50]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[122]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[194]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[266]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[338]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[410]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[482]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 416)] * kernel_shared[554]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[57]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[129]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[201]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[273]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[345]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[417]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[489]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 495)] * kernel_shared[561]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[58]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[130]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[202]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[274]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[346]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[418]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[490]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 496)] * kernel_shared[562]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[59]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[131]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[203]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[275]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[347]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[419]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[491]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 497)] * kernel_shared[563]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[66]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[138]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[210]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[282]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[354]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[426]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[498]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 576)] * kernel_shared[570]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[67]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[139]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[211]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[283]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[355]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[427]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[499]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 577)] * kernel_shared[571]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[68]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[140]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[212]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[284]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[356]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[428]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[500]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 578)] * kernel_shared[572]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[42]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[114]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[186]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[258]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[330]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[402]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[474]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 342)] * kernel_shared[546]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[43]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[115]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[187]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[259]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[331]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[403]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[475]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 343)] * kernel_shared[547]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[44]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[116]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[188]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[260]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[332]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[404]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[476]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 344)] * kernel_shared[548]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[51]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[123]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[195]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[267]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[339]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[411]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[483]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 423)] * kernel_shared[555]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[52]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[124]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[196]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[268]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[340]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[412]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[484]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 424)] * kernel_shared[556]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[53]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[125]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[197]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[269]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[341]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[413]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[485]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 425)] * kernel_shared[557]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[60]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[132]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[204]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[276]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[348]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[420]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[492]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 504)] * kernel_shared[564]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[61]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[133]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[205]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[277]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[349]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[421]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[493]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 505)] * kernel_shared[565]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[62]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[134]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[206]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[278]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[350]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[422]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[494]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 506)] * kernel_shared[566]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[69]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[141]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[213]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[285]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[357]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[429]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[501]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 585)] * kernel_shared[573]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[70]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[142]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[214]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[286]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[358]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[430]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[502]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 586)] * kernel_shared[574]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[71]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[143]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[215]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[287]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[359]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[431]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[503]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((((int)threadIdx.x) / 7) * 9) + (((int)threadIdx.x) % 7)) + 587)] * kernel_shared[575]));
}
- for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
- compute[(((((((int)blockIdx.x) / 7) * 1568) + (((int)threadIdx.x) * 49)) + (i2_inner * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i2_inner] + bias[(((((int)blockIdx.x) / 7) * 32) + ((int)threadIdx.x))]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 8; ++i1_inner) {
+ compute[(((((int)blockIdx.x) * 392) + (i1_inner * 49)) + ((int)threadIdx.x))] = max((conv2d_nchw[i1_inner] + bias[((((int)blockIdx.x) * 8) + i1_inner)]), 0.000000e+00f);
}
}
</pre></div>
@@ -2421,7 +1938,7 @@ In the example below we resume the status and do more 5 trials.</p>
Get devices for measurement successfully!
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 36.299 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 33.658 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 a587a8aee..917a25f17 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -901,7 +901,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 10.1770 10.1816 10.2300 10.1194 0.0453
+ 9.8357 9.8624 9.8666 9.7782 0.0407
</pre></div>
</div>
</div>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
index af236cfdb..29fc5a60e 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -920,7 +920,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 756.4694 756.3174 758.9303 754.1606 1.9502
+ 764.1533 763.5268 766.3735 762.5596 1.6188
</pre></div>
</div>
</div>
@@ -942,7 +942,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 19.969 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 20.635 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 72d43c60a..41bed2aa9 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -620,27 +620,102 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_9: placeholder_15: Buffer(placeholder_14, float32, [128, 512], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_5: placeholder_17: Buffer(placeholder_10, float32, [128, 256], []), placeholder_7: placeholder_18: Buffer(placeholder_12, int32, [4916], []), placeholder_8: placeholder_19: Buffer(placeholder_13, int32, [33], [])} {
- for (i0.outer: int32, 0, 64) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [64]), storage_scope = global;
- for (i1.outer: int32, 0, 16) {
- for (i.outer.inner: int32, 0, 2) {
- for (nb_j.inner: int32, 0, 2) {
- for (j.init: int32, 0, 16) {
- compute_5: Buffer(compute_4, float32, [64], [])[(((i.outer.inner*32) + (nb_j.inner*16)) + j.init)] = 0f32
+ preflattened_buffer_map = {placeholder_5: placeholder_15: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_7: placeholder_18: Buffer(placeholder_12, int32, [4916], []), placeholder_8: placeholder_19: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], [])} {
+ for (i0.outer: int32, 0, 16) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [128]), storage_scope = global;
+ for (i1.outer: int32, 0, 32) {
+ for (i.inner.init: int32, 0, 8) {
+ let cse_var_1: int32 = (i.inner.init*16)
+ {
+ compute_5: Buffer(compute_4, float32, [128], [])[cse_var_1] = 0f32
+ compute_5[(cse_var_1 + 1)] = 0f32
+ compute_5[(cse_var_1 + 2)] = 0f32
+ compute_5[(cse_var_1 + 3)] = 0f32
+ compute_5[(cse_var_1 + 4)] = 0f32
+ compute_5[(cse_var_1 + 5)] = 0f32
+ compute_5[(cse_var_1 + 6)] = 0f32
+ compute_5[(cse_var_1 + 7)] = 0f32
+ compute_5[(cse_var_1 + 8)] = 0f32
+ compute_5[(cse_var_1 + 9)] = 0f32
+ compute_5[(cse_var_1 + 10)] = 0f32
+ compute_5[(cse_var_1 + 11)] = 0f32
+ compute_5[(cse_var_1 + 12)] = 0f32
+ compute_5[(cse_var_1 + 13)] = 0f32
+ compute_5[(cse_var_1 + 14)] = 0f32
+ compute_5[(cse_var_1 + 15)] = 0f32
+ }
+ }
+ for (elem_idx: int32, 0, (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])) {
+ for (i.inner: int32, 0, 8) {
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_2: int32 = (i.inner*16)
+ compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[i1.outer]*16) + (elem_idx*16))]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_3: int32 = ((i.inner*16) + 1)
+ compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_4: int32 = ((i.inner*16) + 2)
+ compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_5: int32 = ((i.inner*16) + 3)
+ compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_6: int32 = ((i.inner*16) + 4)
+ compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_7: int32 = ((i.inner*16) + 5)
+ compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_8: int32 = ((i.inner*16) + 6)
+ compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_9: int32 = ((i.inner*16) + 7)
+ compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_10: int32 = ((i.inner*16) + 8)
+ compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_11: int32 = ((i.inner*16) + 9)
+ compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_12: int32 = ((i.inner*16) + 10)
+ compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_13: int32 = ((i.inner*16) + 11)
+ compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_14: int32 = ((i.inner*16) + 12)
+ compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_15: int32 = ((i.inner*16) + 13)
+ compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_16: int32 = ((i.inner*16) + 14)
+ compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
}
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((i1.outer*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((i1.outer*2) + nb_j.inner)
- let cse_var_2: int32 = (((i.outer.inner*32) + (nb_j.inner*16)) + j)
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((i0.outer*512) + (i.outer.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
- }
+ if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
+ let cse_var_17: int32 = ((i.inner*16) + 15)
+ compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((i0.outer*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
}
}
}
- for (i0.inner: int32, 0, 2) {
- let cse_var_4: int32 = (((i0.outer*1024) + (i0.inner*512)) + (i1.outer*32))
- compute[ramp(cse_var_4, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 8) {
+ let cse_var_18: int32 = (((i0.outer*4096) + (i0.inner*512)) + (i1.outer*16))
+ compute[ramp(cse_var_18, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_18, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -678,7 +753,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.902 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.829 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 4ed19cc88..7d021770c 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -322,7 +322,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:43.177</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:43.701</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -331,22 +331,22 @@
</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>00:43.145</p></td>
+<td><p>00:43.664</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>00:00.019</p></td>
+<td><p>00:00.022</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>
+<tr class="row-odd"><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>00:00.005</p></td>
<td><p>0.0 MB</p></td>
</tr>
-<tr class="row-even"><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>00:00.004</p></td>
+<tr class="row-even"><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>
+<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_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>
+<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>00:00.004</p></td>
<td><p>0.0 MB</p></td>
</tr>
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 797f8f54f..a58a8e960 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -1164,8 +1164,8 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2885496
-No: 6 GFLOPS: 109.11/109.11 result: MeasureResult(costs=(0.002121732916666667,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6635711193084717, timestamp=1656094762.5332358) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
-No: 7 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 6 GFLOPS: 110.60/110.60 result: MeasureResult(costs=(0.0020930916875,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8403575420379639, timestamp=1656095896.9995742) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
+No: 7 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1288,7 +1288,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6225319
-No: 8 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 8 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1411,7 +1411,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,943546
-No: 9 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 9 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1534,7 +1534,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2868708
-No: 10 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 10 GFLOPS: 0.00/110.60 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
@@ -1552,7 +1552,7 @@ No: 10 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
TimeoutError
[('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4691833
-No: 11 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 11 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1675,7 +1675,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1042124
-No: 12 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 12 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1798,7 +1798,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10013405
-No: 13 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 13 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1921,7 +1921,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6732082
-No: 14 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 14 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2044,7 +2044,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7536735
-No: 15 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 15 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2167,7 +2167,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,482121
-No: 16 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 16 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2290,7 +2290,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2824525
-No: 17 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 17 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2413,7 +2413,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4559286
-No: 18 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 18 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2536,7 +2536,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9677544
-No: 19 GFLOPS: 0.00/109.11 result: Traceback (most recent call last):
+No: 19 GFLOPS: 0.00/110.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 738, in __call__
yield remote, remote.load_module(os.path.split(build_result.filename)[1])
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 702, in run_through_rpc
@@ -2624,7 +2624,7 @@ tvm._ffi.base.TVMError: Traceback (most recent call last):
15: _PyEval_EvalFrameDefault
14: 0x0000000000537c30
13: _PyObject_FastCallKeywords
- 12: 0x00007f7e217a3fa2
+ 12: 0x00007fb124dccfa2
11: _ctypes_callproc
10: ffi_call
9: ffi_call_unix64
@@ -2689,7 +2689,7 @@ Traceback (most recent call last):
21: _PyFunction_FastCallKeywords
20: _PyEval_EvalFrameDefault
19: _PyFunction_FastCall [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6390073
-No: 20 GFLOPS: 141.78/141.78 result: MeasureResult(costs=(0.0016327872741935482,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.1234588623046875, timestamp=1656094788.7532756) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
+No: 20 GFLOPS: 144.84/144.84 result: MeasureResult(costs=(0.0015983578700000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4103615283966064, timestamp=1656095923.5313025) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2730,7 +2730,7 @@ and measure running time.</p>
Best config:
[('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
Finish loading 20 records
-Time cost of this operator: 0.002009
+Time cost of this operator: 0.001963
</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 4340a7fb7..2e2af15c0 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -578,10 +578,10 @@ the tuned operator.</p>
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.7 98.727 (1, 2, 10, 10, 3) 2 1
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.106 0.981 (1, 6, 10, 10) 1 1
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.927 0.293 (1, 1, 10, 10, 3) 1 1
-Total_time - 316.733 - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 315.4 98.746 (1, 2, 10, 10, 3) 2 1
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.085 0.966 (1, 6, 10, 10) 1 1
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.921 0.288 (1, 1, 10, 10, 3) 1 1
+Total_time - 319.406 - - - -
</pre></div>
</div>
</div>
@@ -634,10 +634,10 @@ Total_time -
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 192.2 98.556 (1, 1, 10, 10, 6) 2 1
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.982 1.017 (1, 6, 10, 10) 1 1
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.834 0.428 (1, 3, 10, 10, 1) 1 1
-Total_time - 195.017 - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 89.5 97.089 (1, 6, 10, 10, 1) 2 1
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.761 1.911 (1, 6, 10, 10) 1 1
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.922 1.0 (1, 1, 10, 10, 3) 1 1
+Total_time - 92.183 - - - -
</pre></div>
</div>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index 09cb2e691..3394592ed 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -510,7 +510,7 @@ take about <strong>2 minutes</strong> to download the Stanford Cars, while COCO
<a href="https://docs.python.org/3/library/shutil.html#shutil.move" title="shutil.move" class="sphx-glr-backref-module-shutil sphx-glr-backref-type-py-function"><span class="n">shutil</span><span class="o">.</span><span class="n">move</span></a><span class="p">(</span><span class="sa">f</span><span class="s2">"</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-typ [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmpcbeckp09/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmp2ygzw1mt/images/random'
</pre></div>
</div>
</div>
@@ -570,8 +570,8 @@ objects to other stuff? We can display some examples from our datasets using <co
<span class="n">plt</span><span class="o">.</span><span class="n">axis</span><span class="p">(</span><span class="s2">"off"</span><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpcbeckp09/images/target contains 8144 images
-/tmp/tmpcbeckp09/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmp2ygzw1mt/images/target contains 8144 images
+/tmp/tmp2ygzw1mt/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -683,13 +683,13 @@ the time on our validation set).</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Epoch 1/3
-328/328 - 55s - loss: 0.2223 - accuracy: 0.9234 - val_loss: 0.1299 - val_accuracy: 0.9596
+328/328 - 57s - loss: 0.2227 - accuracy: 0.9251 - val_loss: 0.1449 - val_accuracy: 0.9577
Epoch 2/3
-328/328 - 52s - loss: 0.1018 - accuracy: 0.9615 - val_loss: 0.1280 - val_accuracy: 0.9615
+328/328 - 53s - loss: 0.0938 - accuracy: 0.9636 - val_loss: 0.1395 - val_accuracy: 0.9588
Epoch 3/3
-328/328 - 52s - loss: 0.0694 - accuracy: 0.9742 - val_loss: 0.1528 - val_accuracy: 0.9452
+328/328 - 53s - loss: 0.0669 - accuracy: 0.9755 - val_loss: 0.1180 - val_accuracy: 0.9641
-<keras.callbacks.History object at 0x7f675d8a59d0>
+<keras.callbacks.History object at 0x7f87ac6b3c90>
</pre></div>
</div>
</div>
@@ -951,7 +951,7 @@ as intended.</p>
<p>From here, we could modify the model to read live images from the camera - we have another
Arduino tutorial for how to do that <a class="reference external" href="https://github.com/guberti/tvm-arduino-demos/tree/master/examples/person_detection">on GitHub</a>. Alternatively, we could also
<a class="reference external" href="https://tvm.apache.org/docs/how_to/work_with_microtvm/micro_autotune.html">use TVM’s autotuning capabilities</a> to dramatically improve the model’s performance.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 7 minutes 59.583 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 7 minutes 49.189 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 181be64fb..8806d58e3 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -322,7 +322,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>08:45.837</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>08:36.560</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -331,15 +331,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_train.html#sphx-glr-how-to-work-with-microtvm-micro-train-py"><span class="std std-ref">Training Vision Models for microTVM on Arduino</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_train.py</span></code>)</p></td>
-<td><p>07:59.583</p></td>
+<td><p>07:49.189</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>00:42.799</p></td>
+<td><p>00:43.769</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:03.455</p></td>
+<td><p>00:03.601</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index fea554e81..410754f13 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -322,7 +322,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-relay-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:09.539</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:13.029</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -331,11 +331,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><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>00:08.075</p></td>
+<td><p>00:10.510</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><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>00:01.459</p></td>
+<td><p>00:02.513</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_schedules/intrin_math.html b/docs/how_to/work_with_schedules/intrin_math.html
index 150d659f0..ee04b6068 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -515,7 +515,7 @@ 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 0x7f66d2f2fb90>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7f8724cb1170>
</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 b9860fed8..eedf49a15 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -322,7 +322,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:04.018</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:04.304</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -331,35 +331,35 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></td>
-<td><p>00:01.886</p></td>
+<td><p>00:01.902</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></td>
-<td><p>00:00.920</p></td>
+<td><p>00:01.177</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>00:00.524</p></td>
+<td><p>00:00.528</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="scan.html#sphx-glr-how-to-work-with-schedules-scan-py"><span class="std std-ref">Scan and Recurrent Kernel</span></a> (<code class="docutils literal notranslate"><span class="pre">scan.py</span></code>)</p></td>
-<td><p>00:00.517</p></td>
+<td><p>00:00.512</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>00:00.098</p></td>
+<td><p>00:00.105</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></td>
-<td><p>00:00.033</p></td>
+<td><p>00:00.034</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></td>
-<td><p>00:00.028</p></td>
+<td><p>00:00.031</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>00:00.012</p></td>
+<td><p>00:00.013</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index 7bc3114a0..6d6722c28 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -571,7 +571,7 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpt4zzdkfh/input0.cc'\nsource_filename = \"/tmp/tmpt4zzdkfh/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpdpnmy4_h/input0.cc'\nsource_filename = \"/tmp/tmpdpnmy4_h/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index 29ff24c6f..f8469d1b0 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1737,7 +1737,7 @@ Can be the a function or the function name.</p></li>
<dl class="py function">
<dt class="sig sig-object py" id="tvm.auto_scheduler.auto_schedule">
-<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
<dd><p>THIS API IS DEPRECATED.</p>
<p>Run auto scheduling search for a task.</p>
<dl class="field-list simple">
@@ -1774,7 +1774,7 @@ the initial naive schedule (state).</p>
<dl class="py class">
<dt class="sig sig-object py" id="tvm.auto_scheduler.SketchPolicy">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
<dd><p>The search policy that searches in a hierarchical search space defined by sketches.
The policy randomly samples programs from the space defined by sketches and use evolutionary
search to fine-tune them.</p>
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index 97d732375..2ccf96ace 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/f1d30a27b/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -141,7 +141,7 @@
<div class="tsd-signature tsd-kind-icon">bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Uint8Array</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f1d30a27b/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
</ul>
</aside>
</section>
@@ -151,7 +151,7 @@
<div class="tsd-signature tsd-kind-icon">offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f1d30a27b/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
</ul>
</aside>
</section>
@@ -168,7 +168,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f1d30a27b/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">Uint8Array</span></h4>
@@ -185,7 +185,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f1d30a27b/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -202,7 +202,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/f1d30a27b/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/cachedcallstack.html b/docs/reference/api/typedoc/classes/cachedcallstack.html
index 8612fa7a4..39511bea8 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/f1d30a27b/web/src/memory.ts#L223">memory.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L208">memory.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L312">memory.ts:312</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L284">memory.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L388">memory.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L376">memory.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L267">memory.ts:267</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L243">memory.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L321">memory.ts:321</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L252">memory.ts:252</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L359">memory.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L342">memory.ts:342</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L350">memory.ts:350</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L326">memory.ts:326</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L363">memory.ts:363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L346">memory.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/memory.ts#L334">memory.ts:334</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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 adfafa07c..46bdae69e 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/f1d30a27b/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L262">runtime.ts:262</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/f1d30a27b/web/src/runtime.ts#L260">runtime.ts:260</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L260">runtime.ts:260</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/f1d30a27b/web/src/runtime.ts#L258">runtime.ts:258</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L258">runtime.ts:258</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/f1d30a27b/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L262">runtime.ts:262</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/f1d30a27b/web/src/runtime.ts#L279">runtime.ts:279</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L279">runtime.ts:279</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/f1d30a27b/web/src/runtime.ts#L270">runtime.ts:270</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L270">runtime.ts:270</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 a96477d80..c1d293f65 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/f1d30a27b/web/src/runtime.ts#L202">runtime.ts:202</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L202">runtime.ts:202</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/f1d30a27b/web/src/runtime.ts#L200">runtime.ts:200</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L200">runtime.ts:200</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/f1d30a27b/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L198">runtime.ts:198</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/f1d30a27b/web/src/runtime.ts#L223">runtime.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L223">runtime.ts:223</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/f1d30a27b/web/src/runtime.ts#L230">runtime.ts:230</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/runtime.ts#L230">runtime.ts:230</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 8ae87bbab..3f5a151bb 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/f1d30a27b/web/src/environment.ts#L86">environment.ts:86</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/environment.ts#L70">environment.ts:70</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/environment.ts#L69">environment.ts:69</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/environment.ts#L78">environment.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/environment.ts#L84">environment.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/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/f1d30a27b/web/src/environment.ts#L105">environment.ts:105</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/2c3b56c9d/web/src/environment.ts#L105">environment.ts:105</a></li>
</ul>
... 2516 lines suppressed ...