You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2023/03/24 03:50:20 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@e5ae4347ddf492a9e017dd1dba47df364a27a197)
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 113f5bfa35 deploying docs (apache/tvm@e5ae4347ddf492a9e017dd1dba47df364a27a197)
113f5bfa35 is described below
commit 113f5bfa357c79060da693e62de5da2f40e3fd95
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Fri Mar 24 03:50:12 2023 +0000
deploying docs (apache/tvm@e5ae4347ddf492a9e017dd1dba47df364a27a197)
---
docs/_images/sphx_glr_micro_train_001.png | Bin 327396 -> 333957 bytes
docs/_images/sphx_glr_micro_train_thumb.png | Bin 23429 -> 23827 bytes
.../how_to/compile_models/from_darknet.rst.txt | 2 +-
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_oneflow.rst.txt | 2 +-
.../how_to/compile_models/from_pytorch.rst.txt | 2 +-
.../how_to/compile_models/from_tensorflow.rst.txt | 2 +-
.../compile_models/sg_execution_times.rst.txt | 22 +-
.../deploy_models/deploy_model_on_adreno.rst.txt | 2 +-
.../deploy_model_on_adreno_tvmc.rst.txt | 2 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 24 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 2 +-
.../how_to/extend_tvm/sg_execution_times.rst.txt | 10 +-
.../how_to/extend_tvm/use_pass_instrument.rst.txt | 16 +-
.../optimize_operators/opt_conv_cuda.rst.txt | 2 +-
.../optimize_operators/opt_conv_tensorcore.rst.txt | 2 +-
.../how_to/optimize_operators/opt_gemm.rst.txt | 16 +-
.../optimize_operators/sg_execution_times.rst.txt | 8 +-
.../sg_execution_times.rst.txt | 14 +-
.../tune_conv2d_layer_cuda.rst.txt | 1436 ++------------------
.../tune_network_cuda.rst.txt | 4 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 26 +-
.../tune_with_autotvm/sg_execution_times.rst.txt | 8 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 194 ++-
.../work_with_microtvm/micro_autotune.rst.txt | 18 +-
.../work_with_microtvm/micro_pytorch.rst.txt | 4 +-
.../how_to/work_with_microtvm/micro_train.rst.txt | 18 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 14 +-
.../work_with_relay/sg_execution_times.rst.txt | 8 +-
.../how_to/work_with_schedules/intrin_math.rst.txt | 2 +-
.../work_with_schedules/sg_execution_times.rst.txt | 18 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 6 +-
.../frontend/deploy_classification.rst.txt | 2 +-
.../tutorials/frontend/deploy_detection.rst.txt | 2 +-
.../tutorials/frontend/sg_execution_times.rst.txt | 6 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 4 +-
docs/_sources/tutorial/autotvm_matmul_x86.rst.txt | 20 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 61 +-
.../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_darknet.html | 2 +-
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_oneflow.html | 14 +-
docs/how_to/compile_models/from_pytorch.html | 10 +-
docs/how_to/compile_models/from_tensorflow.html | 2 +-
docs/how_to/compile_models/sg_execution_times.html | 22 +-
.../deploy_models/deploy_model_on_adreno.html | 2 +-
.../deploy_models/deploy_model_on_adreno_tvmc.html | 42 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 42 +-
docs/how_to/deploy_models/deploy_prequantized.html | 8 +-
.../deploy_models/deploy_prequantized_tflite.html | 4 +-
docs/how_to/deploy_models/deploy_quantized.html | 2 +-
docs/how_to/deploy_models/deploy_ssd_gluoncv.html | 39 +-
docs/how_to/deploy_models/sg_execution_times.html | 24 +-
.../extend_tvm/bring_your_own_datatypes.html | 2 +-
docs/how_to/extend_tvm/sg_execution_times.html | 10 +-
docs/how_to/extend_tvm/use_pass_instrument.html | 16 +-
docs/how_to/optimize_operators/opt_conv_cuda.html | 2 +-
.../optimize_operators/opt_conv_tensorcore.html | 2 +-
docs/how_to/optimize_operators/opt_gemm.html | 16 +-
.../optimize_operators/sg_execution_times.html | 8 +-
.../sg_execution_times.html | 14 +-
.../tune_conv2d_layer_cuda.html | 1432 ++-----------------
.../tune_with_autoscheduler/tune_network_cuda.html | 4 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 26 +-
.../tune_with_autotvm/sg_execution_times.html | 8 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 194 ++-
docs/how_to/work_with_microtvm/micro_autotune.html | 18 +-
docs/how_to/work_with_microtvm/micro_pytorch.html | 6 +-
docs/how_to/work_with_microtvm/micro_train.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 14 +-
.../how_to/work_with_relay/sg_execution_times.html | 8 +-
docs/how_to/work_with_schedules/intrin_math.html | 2 +-
.../work_with_schedules/sg_execution_times.html | 18 +-
docs/install/nnpack.html | 12 +-
.../reference/api/doxygen/arg__info_8h_source.html | 2 +-
docs/reference/api/doxygen/bytecode_8h_source.html | 2 +-
.../api/doxygen/c__runtime__api_8h_source.html | 2 +-
.../api/doxygen/data__layout_8h_source.html | 2 +-
.../api/doxygen/dataflow__pattern_8h_source.html | 2 +-
.../api/doxygen/detail_2broadcast_8h_source.html | 2 +-
.../api/doxygen/detail_2extern_8h_source.html | 2 +-
.../api/doxygen/device__api_8h_source.html | 2 +-
docs/reference/api/doxygen/elemwise_8h_source.html | 2 +-
.../api/doxygen/group__norm_8h_source.html | 2 +-
.../api/doxygen/index__map_8h_source.html | 2 +-
.../api/doxygen/memory__manager_8h_source.html | 2 +-
.../api/doxygen/namespacemembers_func_l.html | 2 +-
docs/reference/api/doxygen/namespacemembers_l.html | 2 +-
.../api/doxygen/namespacetvm_1_1topi.html | 17 +-
docs/reference/api/doxygen/ndarray_8h_source.html | 2 +-
.../reference/api/doxygen/operation_8h_source.html | 2 +-
.../reference/api/doxygen/profiling_8h_source.html | 2 +-
.../api/doxygen/ravel__unravel_8h_source.html | 2 +-
docs/reference/api/doxygen/search/all_d.js | 2 +-
docs/reference/api/doxygen/search/functions_c.js | 2 +-
.../api/doxygen/shape__tuple_8h_source.html | 2 +-
docs/reference/api/doxygen/tensor_8h_source.html | 2 +-
.../api/doxygen/tensor__type_8h_source.html | 2 +-
docs/reference/api/doxygen/tir_2ir_8h_source.html | 2 +-
docs/reference/api/doxygen/topi_2nn_8h_source.html | 2 +-
docs/reference/api/doxygen/topi_2transform_8h.html | 6 +-
.../api/doxygen/topi_2transform_8h_source.html | 23 +-
docs/reference/api/python/auto_scheduler.html | 4 +-
docs/reference/api/python/topi.html | 3 +-
.../api/typedoc/classes/bytestreamreader.html | 12 +-
.../api/typedoc/classes/cachedcallstack.html | 34 +-
docs/reference/api/typedoc/classes/dldatatype.html | 12 +-
docs/reference/api/typedoc/classes/dldevice.html | 10 +-
.../reference/api/typedoc/classes/environment.html | 12 +-
docs/reference/api/typedoc/classes/ffilibrary.html | 20 +-
docs/reference/api/typedoc/classes/instance.html | 58 +-
docs/reference/api/typedoc/classes/memory.html | 34 +-
docs/reference/api/typedoc/classes/module.html | 10 +-
docs/reference/api/typedoc/classes/ndarray.html | 22 +-
.../api/typedoc/classes/packedfunccell.html | 6 +-
docs/reference/api/typedoc/classes/rpcserver.html | 14 +-
.../api/typedoc/classes/runtimecontext.html | 22 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
docs/reference/api/typedoc/classes/tvmarray.html | 16 +-
docs/reference/api/typedoc/classes/tvmobject.html | 12 +-
.../api/typedoc/classes/webgpucontext.html | 12 +-
docs/reference/api/typedoc/enums/argtypecode.html | 30 +-
.../api/typedoc/enums/aynccallbackcode.html | 4 +-
.../api/typedoc/enums/dldatatypecode.html | 8 +-
.../api/typedoc/enums/rpcserverstate.html | 12 +-
docs/reference/api/typedoc/enums/sizeof.html | 18 +-
docs/reference/api/typedoc/index.html | 124 +-
.../api/typedoc/interfaces/disposable.html | 2 +-
.../api/typedoc/interfaces/functioninfo.html | 6 +-
.../api/typedoc/interfaces/libraryprovider.html | 4 +-
docs/searchindex.js | 2 +-
.../vta/tutorials/autotvm/sg_execution_times.html | 6 +-
.../tutorials/frontend/deploy_classification.html | 2 +-
.../vta/tutorials/frontend/deploy_detection.html | 2 +-
.../vta/tutorials/frontend/sg_execution_times.html | 10 +-
.../vta/tutorials/optimize/sg_execution_times.html | 6 +-
docs/topic/vta/tutorials/sg_execution_times.html | 6 +-
docs/tutorial/auto_scheduler_matmul_x86.html | 4 +-
docs/tutorial/autotvm_matmul_x86.html | 20 +-
docs/tutorial/autotvm_relay_x86.html | 266 ++--
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 +-
159 files changed, 1566 insertions(+), 3587 deletions(-)
diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index 6a9ff83b6f..b0c19c3436 100644
Binary files a/docs/_images/sphx_glr_micro_train_001.png and b/docs/_images/sphx_glr_micro_train_001.png differ
diff --git a/docs/_images/sphx_glr_micro_train_thumb.png b/docs/_images/sphx_glr_micro_train_thumb.png
index faf94be3ac..475ea5ff8a 100644
Binary files a/docs/_images/sphx_glr_micro_train_thumb.png and b/docs/_images/sphx_glr_micro_train_thumb.png differ
diff --git a/docs/_sources/how_to/compile_models/from_darknet.rst.txt b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
index 2c2d4d67e6..132f17aac2 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -318,7 +318,7 @@ The process is no different from other examples.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 19.896 seconds)
+ **Total running time of the script:** ( 1 minutes 23.554 seconds)
.. _sphx_glr_download_how_to_compile_models_from_darknet.py:
diff --git a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
index f0b35eac04..1421bba976 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -116,7 +116,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip0132a9fd-574d-4c94-a24a-11934417a0f0 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip0345609a-aa53-4983-888b-c0a10eb99acb 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 d153fd3d32..a90903e7fd 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -121,7 +121,7 @@ Load a pretrained OneFlow model and save model
.. code-block:: none
Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
-
0%| | 0.00/41.5M [00:00<?, ?B/s]
19%|#9 | 7.99M/41.5M [00:00<00:00, 58.0MB/s]
35%|###4 | 14.3M/41.5M [00:00<00:00, 47.5MB/s]
54%|#####3 | 22.3M/41.5M [00:00<00:00, 58.0MB/s]
68%|######7 | 28.1M/41.5M [00:00<00:00, 53.5MB/s]
80%|######## | 33.3M/41.5M [00:00<00:00, 50.0MB/s]
96%|#########6| 40.0M/41.5M [00:00<00:00, 51.5MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 53.6MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
15%|#5 | 6.33M/41.5M [00:00<00:00, 66.1MB/s]
30%|### | 12.6M/41.5M [00:00<00:00, 51.6MB/s]
43%|####2 | 17.8M/41.5M [00:00<00:00, 33.7MB/s]
58%|#####7 | 24.0M/41.5M [00:00<00:00, 35.9MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 46.0MB/s]
92%|#########2| 38.4M/41.5M [00:00<00:00, 51.2MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 46.5MB/s]
diff --git a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
index ce0ff782fe..61201223eb 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -101,7 +101,7 @@ Load a pretrained PyTorch model
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
0%| | 0.00/44.7M [00:00<?, ?B/s]
27%|##7 | 12.1M/44.7M [00:00<00:00, 127MB/s]
54%|#####4 | 24.2M/44.7M [00:00<00:00, 107MB/s]
77%|#######7 | 34.6M/44.7M [00:00<00:00, 106MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 106MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
18%|#7 | 7.99M/44.7M [00:00<00:00, 76.6MB/s]
40%|#### | 18.1M/44.7M [00:00<00:00, 86.7MB/s]
59%|#####8 | 26.3M/44.7M [00:00<00:00, 79.6MB/s]
76%|#######5 | 33.9M/44.7M [00:00<00:00, 78.7MB/s]
96%|#########5| 42.8M/44.7M [00:00<00:00, 83.6MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 74.4MB/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 545932f891..1c611705a0 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -430,7 +430,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 31.401 seconds)
+ **Total running time of the script:** ( 1 minutes 35.258 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 4dc20d531b..1c5d283a4b 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
=================
-**06:49.101** total execution time for **how_to_compile_models** files:
+**07:04.515** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:31.401 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:35.258 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:19.896 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:23.554 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:57.247 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:59.622 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:38.090 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:39.309 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:32.153 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:33.471 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:30.114 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:31.014 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:28.133 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:28.503 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:26.369 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:27.353 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:22.958 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:23.635 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.738 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.796 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
index c41b98d617..cef636381e 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
@@ -637,7 +637,7 @@ well as provides information about the model's performance
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 2542.2163 2541.2295 2547.7983 2540.4597 2.4130
+ 2750.3544 2750.1791 2754.7262 2747.9918 1.7600
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
index a09e0e9eb8..85f3fe88a7 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
@@ -127,7 +127,7 @@ Make a Keras Resnet50 Model
.. code-block:: none
Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5
-
8192/102967424 [..............................] - ETA: 0s
7725056/102967424 [=>............................] - ETA: 0s
12730368/102967424 [==>...........................] - ETA: 0s
17481728/102967424 [====>.........................] - ETA: 0s
22609920/102967424 [=====>........................] - ETA: 0s
26034176/102967424 [======>.......................] - ETA: 0s
32751616/102967424 [========>.....................] - ETA: 0s
35971072/102967424 [=========>....................] - ETA: 0s
42573824/102967424 [===========>..................] - ETA: 0s
47718400/102967424 [============>.................] - ETA: 0s
50323456/102967424 [=============>................] - ETA: 0s
58023936/102967424 [===============>..............] - ETA: 0s
63528960/102967424 [=================>............] - ETA: 0s
68501504/102967424 [==================>...........] - ETA: 0s
73588736/102967424 [====================>.........] - ETA: 0s
78610432/102967424 [=====================>........] -
ETA: 0s
83623936/102967424 [=======================>......] - ETA: 0s
88727552/102967424 [========================>.....] - ETA: 0s
93495296/102967424 [==========================>...] - ETA: 0s
98705408/102967424 [===========================>..] - ETA: 0s
102967424/102967424 [==============================] - 1s 0us/step
+
8192/102967424 [..............................] - ETA: 0s
4775936/102967424 [>.............................] - ETA: 1s
8380416/102967424 [=>............................] - ETA: 1s
10575872/102967424 [==>...........................] - ETA: 1s
15024128/102967424 [===>..........................] - ETA: 1s
16769024/102967424 [===>..........................] - ETA: 1s
25157632/102967424 [======>.......................] - ETA: 1s
33546240/102967424 [========>.....................] - ETA: 1s
40189952/102967424 [==========>...................] - ETA: 0s
41934848/102967424 [===========>..................] - ETA: 1s
47046656/102967424 [============>.................] - ETA: 0s
50323456/102967424 [=============>................] - ETA: 0s
56967168/102967424 [===============>..............] - ETA: 0s
58712064/102967424 [================>.............] - ETA: 0s
65323008/102967424 [==================>...........] - ETA: 0s
67305472/102967424 [==================>...........] -
ETA: 0s
69296128/102967424 [===================>..........] - ETA: 0s
75489280/102967424 [====================>.........] - ETA: 0s
82878464/102967424 [=======================>......] - ETA: 0s
83877888/102967424 [=======================>......] - ETA: 0s
90521600/102967424 [=========================>....] - ETA: 0s
92266496/102967424 [=========================>....] - ETA: 0s
98910208/102967424 [===========================>..] - ETA: 0s
102850560/102967424
[============================>.] - ETA: 0s
102967424/102967424 [==============================] - 2s 0us/step
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
index dfdfd4290f..128928981f 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
@@ -437,7 +437,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.6415 15.6379 15.7834 15.5298 0.0603
+ 16.1671 16.1450 16.5010 15.9790 0.1270
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 264c794ab0..d59b47dc87 100644
--- a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
@@ -130,7 +130,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MaskRCNN_ResNet50_FPN_Weights.COCO_V1`. You can also use `weights=MaskRCNN_ResNet50_FPN_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
0%| | 0.00/170M [00:00<?, ?B/s]
5%|4 | 8.12M/170M [00:00<00:03, 51.5MB/s]
13%|#3 | 22.6M/170M [00:00<00:01, 81.8MB/s]
18%|#7 | 30.5M/170M [00:00<00:01, 77.6MB/s]
25%|##4 | 42.1M/170M [00:00<00:01, 91.4MB/s]
34%|###3 | 57.7M/170M [00:00<00:01, 114MB/s]
42%|####2 | 72.0M/170M [00:00<00:00, 122MB/s]
49%|####9 | 84.0M/170M [00:00<00:00, 105MB/s]
57%|#####6 | 96.0M/170M [00:00<00:00, 109MB/s]
64%|######4 | 109M/170M [00:01<00:00, 118MB/s]
71%|#######1 | 121M/170M [00:01<00:00, 111MB/s]
78%|#######7 | 132M/170M [00:01<00:00, 105MB/s]
84%|########4 | 143M/170M [00:01<00:00, 107MB/s]
90%|######### | 153M/170M [00:01<00:00, 105MB/s]
96%|#########6| 164M/170M [00:01<00:00, 90.5MB/s]
100%|##########| 170M/170M [00:01<00:00, 101MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
4%|3 | 6.30M/170M [00:00<00:02, 61.9MB/s]
7%|7 | 12.2M/170M [00:00<00:02, 57.3MB/s]
12%|#1 | 19.8M/170M [00:00<00:02, 66.9MB/s]
15%|#5 | 26.2M/170M [00:00<00:02, 53.9MB/s]
19%|#8 | 32.0M/170M [00:00<00:02, 50.7MB/s]
24%|##3 | 40.0M/170M [00:00<00:02, 52.2MB/s]
28%|##8 | 48.0M/170M [00:00<00:02, 55.5MB/s]
33%|###2 | 56.0M/170M [00:01<00:02, 57.5MB/s]
37%|###6 | 62.3M/170M [00:01<00:01, 58.4MB/s]
40%|#### | 68.1M/170M [00:01<00:01, 59.0MB/s]
43%|####3 | 73.8M/170M [00:01<00:01, 54.7MB/s]
47%|####7 | 80.0M/170M [00:01<00:01, 53.6MB/s]
51%|##### | 86.3M/170M [00:01<00:01, 55.2MB/s]
54%|#####3 | 91.6M/170M [00:01<00:01, 47.5MB/s]
57%|#####6 | 96.3M/170M [00:02<00:02, 34.5MB/s]
61%|######1 | 104M/170M [00:02<00:01, 40.7MB/s]
66%|######5 | 112M/170M [00:02<00:01, 43.0MB/s
]
71%|####### | 120M/170M [00:02<00:01, 46.7MB/s]
75%|#######5 | 128M/170M [00:02<00:00, 50.3MB/s]
80%|######## | 136M/170M [00:02<00:00, 54.8MB/s]
85%|########4 | 144M/170M [00:02<00:00, 56.6MB/s]
89%|########9 | 152M/170M [00:03<00:00, 54.0MB/s]
93%|#########3| 158M/170M [00:03<00:00, 55.7MB/s]
97%|#########6| 165M/170M [00:03<00:00, 58.2MB/s]
100%|##########| 170M/170M [00:03<00:00, 51.9MB/s]
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -299,7 +299,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 37.138 seconds)
+ **Total running time of the script:** ( 3 minutes 46.558 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 0c701657ed..2592831a15 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -227,7 +227,7 @@ training. Other models require a full post training calibration.
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MobileNet_V2_Weights.IMAGENET1K_V1`. You can also use `weights=MobileNet_V2_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
-
0%| | 0.00/13.6M [00:00<?, ?B/s]
92%|#########2| 12.5M/13.6M [00:00<00:00, 131MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 135MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
59%|#####8 | 7.99M/13.6M [00:00<00:00, 38.6MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 58.9MB/s]
@@ -409,7 +409,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.7204 90.6843 92.1979 90.2426 0.3522
+ 88.9308 88.2800 91.3655 88.0628 0.9692
@@ -458,7 +458,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 17.958 seconds)
+ **Total running time of the script:** ( 1 minutes 19.728 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 108f6d6751..5c779b4616 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
@@ -423,7 +423,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 121.0102 120.9625 123.1915 119.6895 0.5799
+ 121.0844 121.2201 122.4309 118.6074 0.7515
@@ -460,7 +460,7 @@ Here we give an example of how to measure performance of TVM compiled models.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 35.394 seconds)
+ **Total running time of the script:** ( 2 minutes 29.938 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 3af9e6c456..08de27c30d 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -257,7 +257,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 35.606 seconds)
+ **Total running time of the script:** ( 1 minutes 39.936 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 7fda66571f..0b35540685 100644
--- a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
@@ -170,7 +170,7 @@ Convert and compile model for CPU.
data: None
input_sym_arg_type = in_param.infer_type()[0]
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
-
0%| | 0/132723 [00:00<?, ?KB/s]
1%| | 1194/132723 [00:00<00:11, 11915.01KB/s]
3%|2 | 3786/132723 [00:00<00:06, 20058.73KB/s]
7%|7 | 9509/132723 [00:00<00:03, 36981.11KB/s]
13%|#3 | 17878/132723 [00:00<00:02, 55386.06KB/s]
20%|#9 | 26291/132723 [00:00<00:01, 65737.89KB/s]
26%|##6 | 34665/132723 [00:00<00:01, 71850.50KB/s]
33%|###2 | 43164/132723 [00:00<00:01, 76142.21KB/s]
39%|###8 | 51232/132723 [00:00<00:01, 77583.45KB/s]
45%|####5 | 59733/132723 [00:00<00:00, 79902.00KB/s]
51%|#####1 | 68271/132723 [00:01<00:00, 81589.82KB/s]
58%|#####7 | 76639/132723 [00:01<00:00, 82227.57KB/s]
64%|######4 | 85076/132723 [00:01<00:00, 82877.69KB/s]
70%|####### | 93520/132723 [00:01<00:00, 83347.81KB/s]
77%|#######6 | 101994/132723 [00:01<00:00, 83766.91KB/s]
83%|########3 | 110500/132723 [00:01<00:00, 84154.44KB/s]
90%|########9 |
118916/132723 [00:01<00:00, 84030.66KB/s]
96%|#########5| 127413/132723 [00:01<00:00, 84309.45KB/s]
100%|##########| 132723/132723 [00:01<00:00, 75173.46KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
1%| | 1308/132723 [00:00<00:10, 13040.62KB/s]
3%|3 | 4237/132723 [00:00<00:05, 22586.65KB/s]
8%|7 | 10121/132723 [00:00<00:03, 39126.63KB/s]
12%|#2 | 16441/132723 [00:00<00:02, 48624.36KB/s]
18%|#7 | 23636/132723 [00:00<00:01, 57032.13KB/s]
23%|##3 | 31133/132723 [00:00<00:01, 63125.25KB/s]
29%|##9 | 39038/132723 [00:00<00:01, 68327.34KB/s]
35%|###5 | 46987/132723 [00:00<00:01, 71876.62KB/s]
41%|####1 | 54929/132723 [00:00<00:01, 74232.14KB/s]
47%|####7 | 62701/132723 [00:01<00:00, 75299.27KB/s]
53%|#####3 | 70589/132723 [00:01<00:00, 76390.87KB/s]
59%|#####9 | 78560/132723 [00:01<00:00, 77398.99KB/s]
65%|######5 | 86458/132723 [00:01<00:00, 77874.69KB/s]
71%|#######1 | 94448/132723 [00:01<00:00, 78484.07KB/s]
77%|#######7 | 102380/132723 [00:01<00:00, 78735.08KB/s]
83%|########3 |
110273/132723 [00:01<00:00, 78790.94KB/s]
89%|########9 | 118255/132723 [00:01<00:00, 79098.68KB/s]
95%|#########5| 126172/132723 [00:01<00:00, 79118.26KB/s]
100%|##########| 132723/132723 [00:01<00:00, 70493.20KB/s]
@@ -246,7 +246,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 51.147 seconds)
+ **Total running time of the script:** ( 3 minutes 59.099 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 735cc182bf..851f61871e 100644
--- a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
@@ -5,28 +5,28 @@
Computation times
=================
-**16:23.800** total execution time for **how_to_deploy_models** files:
+**16:49.430** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 03:51.147 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 03:59.099 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:37.138 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:46.558 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:35.394 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:29.938 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:35.606 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:39.936 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:17.958 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:19.728 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 00:54.873 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 00:58.394 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``) | 00:51.262 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``) | 00:53.202 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:42.726 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:44.689 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:28.962 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:29.210 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:28.724 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:28.670 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``) | 00:00.010 | 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 38fb86bf25..1bf33b1e0b 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.zipb5ee551d-6d6e-49f4-9bed-cc3ed7b09e10 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip929e5e9d-8d88-4ea5-95fc-5c8e6ad4ee23 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
diff --git a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
index 3aa5f0e562..c5d02f6a30 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
=================
-**00:54.081** total execution time for **how_to_extend_tvm** files:
+**00:58.769** 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``) | 00:50.293 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:54.661 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.725 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.952 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.056 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.149 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.007 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.008 | 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 d79c5b740c..9f241b2239 100644
--- a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
@@ -220,10 +220,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 22339us [22339us] (48.07%; 48.07%)
- FoldScaleAxis: 24132us [7us] (51.93%; 51.93%)
- FoldConstant: 24125us [1720us] (51.91%; 99.97%)
- InferType: 22405us [22405us] (48.21%; 92.87%)
+ InferType: 23565us [23565us] (49.07%; 49.07%)
+ FoldScaleAxis: 24454us [9us] (50.93%; 50.93%)
+ FoldConstant: 24444us [1762us] (50.91%; 99.96%)
+ InferType: 22683us [22683us] (47.24%; 92.79%)
@@ -262,10 +262,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 21709us [21709us] (48.40%; 48.40%)
- FoldScaleAxis: 23141us [5us] (51.60%; 51.60%)
- FoldConstant: 23136us [1729us] (51.59%; 99.98%)
- InferType: 21407us [21407us] (47.73%; 92.53%)
+ InferType: 23804us [23804us] (49.22%; 49.22%)
+ FoldScaleAxis: 24559us [8us] (50.78%; 50.78%)
+ FoldConstant: 24551us [1875us] (50.76%; 99.97%)
+ InferType: 22675us [22675us] (46.89%; 92.36%)
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 0ca1d9fd81..f59edd18d2 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -331,7 +331,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 53.507263 ms
+ Convolution: 53.512542 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 d4f6fe6cf1..216860cd2d 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
@@ -598,7 +598,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 11.759011 ms
+ conv2d with tensor core: 12.223469 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 e89ad8232a..da859c2f73 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -134,8 +134,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.018365
- Baseline: 3.399232
+ Numpy running time: 0.019328
+ Baseline: 3.498159
@@ -227,7 +227,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.299479
+ Opt1: 0.329814
@@ -318,7 +318,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.334021
+ Opt2: 0.353135
@@ -406,7 +406,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.117414
+ Opt3: 0.134273
@@ -523,7 +523,7 @@ flattening.
.. code-block:: none
- Opt4: 0.109954
+ Opt4: 0.110898
@@ -635,7 +635,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.111315
+ Opt5: 0.112132
@@ -748,7 +748,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
.. code-block:: none
- Opt6: 0.147122
+ Opt6: 0.148916
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 a4ebecba3b..4e59eb4bbb 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
=================
-**00:35.456** total execution time for **how_to_optimize_operators** files:
+**00:36.796** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.400 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:33.689 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.930 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.894 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.127 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.212 | 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 9e72d9b713..09ebb8bb55 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
=================
-**10:20.424** total execution time for **how_to_tune_with_autoscheduler** files:
+**10:36.831** total execution time for **how_to_tune_with_autoscheduler** files:
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:25.850 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:28.831 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:42.813 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:47.644 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:10.653 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:13.908 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:33.748 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:36.900 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:13.971 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:15.119 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:13.389 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:14.429 | 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 5d7f6d1353..987f55e14e 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
@@ -209,13 +209,6 @@ file and apply it.
-.. rst-class:: sphx-glr-script-out
-
- .. code-block:: none
-
- .T
-
-
@@ -250,664 +243,86 @@ cooperative fetching, unrolling and operator fusion.
@T.prim_func
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True})
- blockIdx_x = T.launch_thread("blockIdx.x", 16)
- conv2d_nchw = T.allocate([28], "float32", "local")
- pad_temp_shared = T.allocate([1296], "float32", "shared")
- kernel_shared = T.allocate([4608], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 56)
- conv2d_nchw_1 = T.Buffer((16,), data=conv2d_nchw, scope="local", align=16)
+ blockIdx_x = T.launch_thread("blockIdx.x", 32)
+ conv2d_nchw = T.allocate([7], "float32", "local")
+ pad_temp_shared = T.allocate([1568], "float32", "shared")
+ kernel_shared = T.allocate([512], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 112)
+ conv2d_nchw_1 = T.Buffer((7,), data=conv2d_nchw, scope="local", align=16)
conv2d_nchw_1[0] = T.float32(0)
- conv2d_nchw_1[4] = T.float32(0)
- conv2d_nchw_1[8] = T.float32(0)
- conv2d_nchw_1[12] = T.float32(0)
- conv2d_nchw_1[16] = T.float32(0)
- conv2d_nchw_1[20] = T.float32(0)
- conv2d_nchw_1[24] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
- conv2d_nchw_1[5] = T.float32(0)
- conv2d_nchw_1[9] = T.float32(0)
- conv2d_nchw_1[13] = T.float32(0)
- conv2d_nchw_1[17] = T.float32(0)
- conv2d_nchw_1[21] = T.float32(0)
- conv2d_nchw_1[25] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
- conv2d_nchw_1[6] = T.float32(0)
- conv2d_nchw_1[10] = T.float32(0)
- conv2d_nchw_1[14] = T.float32(0)
- conv2d_nchw_1[18] = T.float32(0)
- conv2d_nchw_1[22] = T.float32(0)
- conv2d_nchw_1[26] = T.float32(0)
conv2d_nchw_1[3] = T.float32(0)
- conv2d_nchw_1[7] = T.float32(0)
- conv2d_nchw_1[11] = T.float32(0)
- conv2d_nchw_1[15] = T.float32(0)
- conv2d_nchw_1[19] = T.float32(0)
- conv2d_nchw_1[23] = T.float32(0)
- conv2d_nchw_1[27] = T.float32(0)
- for rc_outer_outer in range(32):
- cse_var_2: T.int32 = rc_outer_outer * 784
- cse_var_1: T.int32 = rc_outer_outer * 144
+ conv2d_nchw_1[4] = T.float32(0)
+ conv2d_nchw_1[5] = T.float32(0)
+ conv2d_nchw_1[6] = T.float32(0)
+ for rc_outer_outer, ry_outer_outer, rx_outer_outer in T.grid(16, 3, 3):
+ cse_var_2: T.int32 = rc_outer_outer * 288
+ cse_var_1: T.int32 = ry_outer_outer * 3
threadIdx_x_1 = T.env_thread("threadIdx.x")
- pad_temp_shared_1 = T.Buffer((1296,), data=pad_temp_shared, scope="shared")
+ pad_temp_shared_1 = T.Buffer((1568,), data=pad_temp_shared, scope="shared")
data_1 = T.Buffer((25088,), data=data.data)
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(9 <= threadIdx_x_1 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 // 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 56] = T.if_then_else(9 <= (threadIdx_x_1 + 56) % 81 and (threadIdx_x_1 + 56) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 56) // 81 * 49 + (threadIdx_x_1 + 56) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 112] = T.if_then_else(9 <= (threadIdx_x_1 + 31) % 81 and (threadIdx_x_1 + 31) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 112) // 81 * 49 + (threadIdx_x_1 + 31) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 168] = T.if_then_else(3 <= threadIdx_x_1 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 168) // 81 * 49 + (threadIdx_x_1 + 6) // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 224] = T.if_then_else(9 <= (threadIdx_x_1 + 62) % 81 and (threadIdx_x_1 + 62) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 224) // 81 * 49 + (threadIdx_x_1 + 62) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 280] = T.if_then_else(9 <= (threadIdx_x_1 + 37) % 81 and (threadIdx_x_1 + 37) % 81 < 72 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 280) // 81 * 49 + (threadIdx_x_1 + 37) % 81 // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 336] = T.if_then_else(1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 336) // 81 * 49 + (threadIdx_x_1 + 12) // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 392] = T.if_then_else(9 <= (threadIdx_x_1 + 68) % 81 and (threadIdx_x_1 + 68) % 81 < 72 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 392) // 81 * 49 + (threadIdx_x_1 + 68) % 81 // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 448] = T.if_then_else(9 <= (threadIdx_x_1 + 43) % 81 and (threadIdx_x_1 + 43) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 448) // 81 * 49 + (threadIdx_x_1 + 43) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 504] = T.if_then_else(threadIdx_x_1 < 54 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 504) // 81 * 49 + threadIdx_x_1 // 9 * 7 + threadIdx_x_1 % 9 + 6], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 560] = T.if_then_else(9 <= (threadIdx_x_1 + 74) % 81 and (threadIdx_x_1 + 74) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 560) // 81 * 49 + (threadIdx_x_1 + 74) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 616] = T.if_then_else(9 <= (threadIdx_x_1 + 49) % 81 and (threadIdx_x_1 + 49) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 616) // 81 * 49 + (threadIdx_x_1 + 49) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 672] = T.if_then_else(threadIdx_x_1 < 48 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 672) // 81 * 49 + (threadIdx_x_1 + 24) // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 728] = T.if_then_else(9 <= (threadIdx_x_1 + 80) % 81 and (threadIdx_x_1 + 80) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 728) // 81 * 49 + (threadIdx_x_1 + 80) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 784] = T.if_then_else(9 <= (threadIdx_x_1 + 55) % 81 and (threadIdx_x_1 + 55) % 81 < 72 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 784) // 81 * 49 + (threadIdx_x_1 + 55) % 81 // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 840] = T.if_then_else(9 <= (threadIdx_x_1 + 30) % 81 and (threadIdx_x_1 + 30) % 81 < 72 and 1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 840) // 81 * 49 + (threadIdx_x_1 + 30) % 81 // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 896] = T.if_then_else(4 <= threadIdx_x_1 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 896) // 81 * 49 + (threadIdx_x_1 + 5) // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 952] = T.if_then_else(9 <= (threadIdx_x_1 + 61) % 81 and (threadIdx_x_1 + 61) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 952) // 81 * 49 + (threadIdx_x_1 + 61) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1008] = T.if_then_else(1 <= (threadIdx_x_1 // 9 + 4) % 9 and (threadIdx_x_1 + 36) % 81 < 72 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1008) // 81 * 49 + (threadIdx_x_1 // 9 + 4) % 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1064] = T.if_then_else(1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1064) // 81 * 49 + (threadIdx_x_1 + 11) // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1120] = T.if_then_else(9 <= (threadIdx_x_1 + 67) % 81 and (threadIdx_x_1 + 67) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1120) // 81 * 49 + (threadIdx_x_1 + 67) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1176] = T.if_then_else(9 <= (threadIdx_x_1 + 42) % 81 and (threadIdx_x_1 + 42) % 81 < 72 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1176) // 81 * 49 + (threadIdx_x_1 + 42) % 81 // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1232] = T.if_then_else(threadIdx_x_1 < 55 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1232) // 81 * 49 + (threadIdx_x_1 + 17) // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- if T.likely(threadIdx_x_1 < 8):
- pad_temp_shared_1[threadIdx_x_1 + 1288] = T.float32(0)
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(1 <= threadIdx_x_1 % 49 // 7 + ry_outer_outer and threadIdx_x_1 % 49 // 7 + ry_outer_outer < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 112] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 2) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 2) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 104], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 224] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 4) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 4) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 216], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 336] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 6) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 6) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 328], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 448] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 1) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 1) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 440], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 560] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 3) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 3) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 552], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 672] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 5) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 5) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 664], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 784] = T.if_then_else(1 <= threadIdx_x_1 % 49 // 7 + ry_outer_outer and threadIdx_x_1 % 49 // 7 + ry_outer_outer < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 776], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 896] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 2) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 2) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 888], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1008] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 4) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 4) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1000], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1120] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 6) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 6) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1112], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1232] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 1) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 1) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1224], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1344] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 3) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 3) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1336], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1456] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 5) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 5) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1448], T.float32(0))
threadIdx_x_2 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((4608,), data=kernel_shared, scope="shared")
+ kernel_shared_1 = T.Buffer((512,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8]
- kernel_shared_1[threadIdx_x_2 * 8 + 1] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 1]
- kernel_shared_1[threadIdx_x_2 * 8 + 2] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 2]
- kernel_shared_1[threadIdx_x_2 * 8 + 3] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 3]
- kernel_shared_1[threadIdx_x_2 * 8 + 4] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 4]
- kernel_shared_1[threadIdx_x_2 * 8 + 5] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 5]
- kernel_shared_1[threadIdx_x_2 * 8 + 6] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 6]
- kernel_shared_1[threadIdx_x_2 * 8 + 7] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 7]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 448] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 16) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 449] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 17) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 450] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 18) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 451] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 19) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 452] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 20) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 453] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 21) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 454] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 22) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 455] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 23) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 896] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 32) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 897] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 33) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 898] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 34) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 899] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 35) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 900] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 36) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 901] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 37) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 902] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 38) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 903] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 39) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 1344] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 48) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1345] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 49) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1346] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 50) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1347] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 51) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1348] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 52) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1349] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 53) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1350] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 54) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1351] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 55) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 1792] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 64) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1793] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 65) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1794] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 66) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1795] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 67) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1796] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 68) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1797] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 69) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1798] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 70) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1799] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 71) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 2240] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 80) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2241] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 81) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2242] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 82) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2243] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 83) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2244] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 84) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2245] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 85) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2246] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 86) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2247] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 87) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 2688] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 96) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2689] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 97) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2690] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 98) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2691] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 99) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2692] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 100) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2693] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 101) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2694] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 102) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2695] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 103) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 3136] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 112) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3137] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 113) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3138] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 114) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3139] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 115) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3140] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 116) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3141] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 117) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3142] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 118) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3143] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 119) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 3584] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 128) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3585] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 129) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3586] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 130) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3587] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 131) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3588] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 132) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3589] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 133) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3590] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 134) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3591] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 135) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 4032] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129024]
- kernel_shared_1[threadIdx_x_2 * 8 + 4033] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129025]
- kernel_shared_1[threadIdx_x_2 * 8 + 4034] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129026]
- kernel_shared_1[threadIdx_x_2 * 8 + 4035] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129027]
- kernel_shared_1[threadIdx_x_2 * 8 + 4036] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129028]
- kernel_shared_1[threadIdx_x_2 * 8 + 4037] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129029]
- kernel_shared_1[threadIdx_x_2 * 8 + 4038] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129030]
- kernel_shared_1[threadIdx_x_2 * 8 + 4039] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129031]
- with T.launch_thread(threadIdx_x_2, 56):
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4480] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 16]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4481] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 17]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4482] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 18]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4483] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 19]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4484] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 20]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4485] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 21]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4486] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 22]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4487] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 23]
- for ry_outer_inner, rx_outer_inner in T.grid(3, 3):
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 3] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 4] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 5] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 6] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 3] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 4] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 5] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 6] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 3] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 4] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 5] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 6] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 3] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 4] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 5] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 6] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 84] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 85] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 86] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 87] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 84] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 85] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 86] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 87] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 84] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 85] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 86] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 87] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 84] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 85] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 86] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 87] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 165] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 166] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 167] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 168] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 165] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 166] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 167] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 168] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 165] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 166] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 167] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 168] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 165] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 166] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 167] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 168] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 246] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 247] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 248] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 249] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 246] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 247] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 248] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 249] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 246] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 247] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 248] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 249] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 246] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 247] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 248] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 249] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 324] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 325] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 326] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 327] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 328] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 329] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 330] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 324] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 325] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 326] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 327] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 328] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 329] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 330] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 324] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 325] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 326] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 327] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 328] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 329] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 330] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 324] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 325] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 326] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 327] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 328] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 329] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 330] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 405] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 406] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 407] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 408] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 409] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 410] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 411] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 405] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 406] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 407] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 408] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 409] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 410] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 411] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 405] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 406] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 407] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 408] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 409] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 410] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 411] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 405] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 406] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 407] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 408] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 409] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 410] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 411] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 486] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 487] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 488] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 489] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 490] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 491] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 492] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 486] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 487] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 488] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 489] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 490] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 491] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 492] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 486] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 487] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 488] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 489] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 490] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 491] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 492] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 486] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 487] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 488] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 489] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 490] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 491] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 492] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 567] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 568] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 569] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 570] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 571] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 572] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 573] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 567] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 568] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 569] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 570] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 571] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 572] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 573] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 567] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 568] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 569] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 570] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 571] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 572] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 573] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 567] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 568] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 569] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 570] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 571] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 572] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 573] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 648] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 649] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 650] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 651] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 652] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 653] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 654] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 648] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 649] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 650] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 651] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 652] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 653] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 654] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 648] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 649] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 650] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 651] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 652] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 653] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 654] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 648] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 649] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 650] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 651] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 652] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 653] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 654] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 729] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 730] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 731] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 732] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 733] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 734] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 735] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 729] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 730] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 731] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 732] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 733] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 734] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 735] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 729] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 730] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 731] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 732] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 733] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 734] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 735] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 729] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 730] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 731] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 732] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 733] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 734] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 735] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 810] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 811] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 812] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 813] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 814] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 815] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 816] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 810] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 811] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 812] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 813] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 814] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 815] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 816] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 810] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 811] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 812] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 813] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 814] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 815] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 816] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 810] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 811] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 812] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 813] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 814] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 815] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 816] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 891] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 892] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 893] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 894] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 895] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 896] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 897] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 891] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 892] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 893] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 894] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 895] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 896] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 897] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 891] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 892] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 893] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 894] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 895] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 896] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 897] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 891] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 892] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 893] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 894] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 895] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 896] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 897] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 972] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 973] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 974] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 975] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 976] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 977] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 978] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 972] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 973] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 974] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 975] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 976] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 977] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 978] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 972] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 973] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 974] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 975] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 976] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 977] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 978] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 972] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 973] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 974] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 975] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 976] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 977] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 978] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1053] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1054] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1055] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1056] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1057] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1058] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1059] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1053] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1054] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1055] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1056] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1057] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1058] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1059] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1053] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1054] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1055] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1056] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1057] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1058] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1059] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1053] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1054] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1055] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1056] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1057] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1058] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1059] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1134] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1135] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1136] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1137] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1138] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1139] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1140] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1134] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1135] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1136] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1137] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1138] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1139] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1140] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1134] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1135] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1136] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1137] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1138] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1139] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1140] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1134] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1135] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1136] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1137] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1138] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1139] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1140] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1215] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1216] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1217] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1218] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1219] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1220] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1221] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1215] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1216] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1217] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1218] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1219] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1220] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1221] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1215] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1216] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1217] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1218] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1219] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1220] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1221] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1215] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1216] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1217] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1218] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1219] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1220] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1221] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- for i1_inner in range(4):
+ with T.launch_thread(threadIdx_x_2, 112):
+ kernel_shared_1[threadIdx_x_2 * 3] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 * 3 // 32 * 4608 + cse_var_2 + threadIdx_x_2 * 3 % 32 * 9 + cse_var_1 + rx_outer_outer]
+ kernel_shared_1[threadIdx_x_2 * 3 + 1] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 1) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 1) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ kernel_shared_1[threadIdx_x_2 * 3 + 2] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 2) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 2) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_2, 112):
+ if T.likely(threadIdx_x_2 < 59):
+ kernel_shared_1[threadIdx_x_2 * 3 + 336] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 336) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 16) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ if T.likely(threadIdx_x_2 < 59):
+ kernel_shared_1[threadIdx_x_2 * 3 + 337] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 337) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 17) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ if T.likely(threadIdx_x_2 < 58):
+ kernel_shared_1[threadIdx_x_2 * 3 + 338] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 338) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 18) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ for rc_outer_inner in range(16):
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 3] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 4] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 5] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 6] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 49] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 50] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 51] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 52] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 53] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ for i3_inner in range(7):
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7] = T.max(conv2d_nchw_1[i1_inner] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 1] = T.max(conv2d_nchw_1[i1_inner + 4] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 2] = T.max(conv2d_nchw_1[i1_inner + 8] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 3] = T.max(conv2d_nchw_1[i1_inner + 12] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 4] = T.max(conv2d_nchw_1[i1_inner + 16] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 5] = T.max(conv2d_nchw_1[i1_inner + 20] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 6] = T.max(conv2d_nchw_1[i1_inner + 24] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 784 + threadIdx_x * 7 + i3_inner] = T.max(conv2d_nchw_1[i3_inner] + bias_1[blockIdx_x * 16 + threadIdx_x // 7], T.float32(0))
@@ -957,7 +372,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.265 ms
+ Execution time of this operator: 0.387 ms
@@ -1005,37 +420,37 @@ 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=4)
+ conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
- conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=8)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=16)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=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_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=7)
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_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=7)
- conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=16)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=1)
+ conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
+ conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=16)
conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
- conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
+ conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
- conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
+ conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
- compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=4)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=8)
+ compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
+ compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=16)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_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_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
- compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_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)
kernel_shared = s.cache_read(kernel, "shared", [conv2d_nchw])
@@ -1052,16 +467,16 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused = s[compute].fuse(compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i)
s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread_axis("threadIdx.x"))
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=8)
+ 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=3)
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=56)
+ kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
- s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+ s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 64)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
CUDA source code:
@@ -1079,632 +494,69 @@ 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__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[28];
- __shared__ float pad_temp_shared[1296];
- __shared__ float kernel_shared[4608];
+ extern "C" __global__ void __launch_bounds__(112) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[7];
+ __shared__ float pad_temp_shared[1568];
+ __shared__ float kernel_shared[512];
conv2d_nchw[0] = 0.000000e+00f;
- conv2d_nchw[4] = 0.000000e+00f;
- conv2d_nchw[8] = 0.000000e+00f;
- conv2d_nchw[12] = 0.000000e+00f;
- conv2d_nchw[16] = 0.000000e+00f;
- conv2d_nchw[20] = 0.000000e+00f;
- conv2d_nchw[24] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
- conv2d_nchw[5] = 0.000000e+00f;
- conv2d_nchw[9] = 0.000000e+00f;
- conv2d_nchw[13] = 0.000000e+00f;
- conv2d_nchw[17] = 0.000000e+00f;
- conv2d_nchw[21] = 0.000000e+00f;
- conv2d_nchw[25] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
- conv2d_nchw[6] = 0.000000e+00f;
- conv2d_nchw[10] = 0.000000e+00f;
- conv2d_nchw[14] = 0.000000e+00f;
- conv2d_nchw[18] = 0.000000e+00f;
- conv2d_nchw[22] = 0.000000e+00f;
- conv2d_nchw[26] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
- conv2d_nchw[7] = 0.000000e+00f;
- conv2d_nchw[11] = 0.000000e+00f;
- conv2d_nchw[15] = 0.000000e+00f;
- conv2d_nchw[19] = 0.000000e+00f;
- conv2d_nchw[23] = 0.000000e+00f;
- conv2d_nchw[27] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 32; ++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 * 784) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 56)] = (((((9 <= ((((int)threadIdx.x) + 56) % 81)) && (((((int)threadIdx.x) + 56) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 56) / 81) * 49)) + ((((((int)threadIdx.x) + 56) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((9 <= ((((int)threadIdx.x) + 31) % 81)) && (((((int)threadIdx.x) + 31) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 112) / 81) * 49)) + ((((((int)threadIdx.x) + 31) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 168)] = ((((3 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 168) / 81) * 49)) + (((((int)threadIdx.x) + 6) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((9 <= ((((int)threadIdx.x) + 62) % 81)) && (((((int)threadIdx.x) + 62) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 224) / 81) * 49)) + ((((((int)threadIdx.x) + 62) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 280)] = (((((9 <= ((((int)threadIdx.x) + 37) % 81)) && (((((int)threadIdx.x) + 37) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 280) / 81) * 49)) + ((((((int)threadIdx.x) + 37) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 336)] = (((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 336) / 81) * 49)) + (((((int)threadIdx.x) + 12) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 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 * 784) + (((((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) + 448)] = (((((9 <= ((((int)threadIdx.x) + 43) % 81)) && (((((int)threadIdx.x) + 43) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 448) / 81) * 49)) + ((((((int)threadIdx.x) + 43) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 504)] = ((((((int)threadIdx.x) < 54) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 504) / 81) * 49)) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) + 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 560)] = (((((9 <= ((((int)threadIdx.x) + 74) % 81)) && (((((int)threadIdx.x) + 74) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 560) / 81) * 49)) + ((((((int)threadIdx.x) + 74) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 616)] = (((((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 * 784) + (((((int)threadIdx.x) + 616) / 81) * 49)) + ((((((int)threadIdx.x) + 49) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 672)] = ((((((int)threadIdx.x) < 48) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 672) / 81) * 49)) + (((((int)threadIdx.x) + 24) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 728)] = (((((9 <= ((((int)threadIdx.x) + 80) % 81)) && (((((int)threadIdx.x) + 80) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 728) / 81) * 49)) + ((((((int)threadIdx.x) + 80) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((9 <= ((((int)threadIdx.x) + 55) % 81)) && (((((int)threadIdx.x) + 55) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 784) / 81) * 49)) + ((((((int)threadIdx.x) + 55) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 840)] = (((((9 <= ((((int)threadIdx.x) + 30) % 81)) && (((((int)threadIdx.x) + 30) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 840) / 81) * 49)) + ((((((int)threadIdx.x) + 30) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 896)] = ((((4 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 896) / 81) * 49)) + (((((int)threadIdx.x) + 5) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 952)] = (((((9 <= ((((int)threadIdx.x) + 61) % 81)) && (((((int)threadIdx.x) + 61) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 952) / 81) * 49)) + ((((((int)threadIdx.x) + 61) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1008)] = (((((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 * 784) + (((((int)threadIdx.x) + 1008) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 4) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1064)] = (((1 <= ((((int)threadIdx.x) + 2) % 9)) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1064) / 81) * 49)) + (((((int)threadIdx.x) + 11) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1120)] = (((((9 <= ((((int)threadIdx.x) + 67) % 81)) && (((((int)threadIdx.x) + 67) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1120) / 81) * 49)) + ((((((int)threadIdx.x) + 67) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((9 <= ((((int)threadIdx.x) + 42) % 81)) && (((((int)threadIdx.x) + 42) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1176) / 81) * 49)) + ((((((int)threadIdx.x) + 42) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1232)] = ((((((int)threadIdx.x) < 55) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1232) / 81) * 49)) + (((((int)threadIdx.x) + 17) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 8) {
- pad_temp_shared[(((int)threadIdx.x) + 1288)] = 0.000000e+00f;
- }
- kernel_shared[(((int)threadIdx.x) * 8)] = kernel[((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 1)];
- kernel_shared[((((int)threadIdx.x) * 8) + 2)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 2)];
- kernel_shared[((((int)threadIdx.x) * 8) + 3)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 3)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 4)];
- kernel_shared[((((int)threadIdx.x) * 8) + 5)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 5)];
- kernel_shared[((((int)threadIdx.x) * 8) + 6)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 6)];
- kernel_shared[((((int)threadIdx.x) * 8) + 7)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 7)];
- kernel_shared[((((int)threadIdx.x) * 8) + 448)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 16) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 449)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 17) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 450)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 18) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 451)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 19) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 452)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 20) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 453)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 21) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 454)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 22) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 455)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 23) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 896)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 32) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 897)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 33) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 898)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 34) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 899)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 35) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 900)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 36) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 901)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 37) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 902)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 38) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 903)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 39) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1344)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 48) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1345)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 49) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1346)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 50) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1347)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 51) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1348)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 52) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1349)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 53) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1350)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 54) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1351)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 55) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1792)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 64) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1793)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 65) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1794)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 66) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1795)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 67) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1796)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 68) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1797)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 69) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1798)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 70) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1799)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 71) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2240)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 80) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2241)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 81) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2242)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 82) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2243)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 83) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2244)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 84) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2245)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 85) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2246)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 86) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2247)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 87) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2688)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 96) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2689)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 97) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2690)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 98) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2691)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 99) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2692)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 100) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2693)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 101) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2694)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 102) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2695)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 103) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3136)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 112) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3137)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 113) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3138)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 114) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3139)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 115) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3140)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 116) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3141)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 117) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3142)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 118) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3143)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 119) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3584)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 128) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3585)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 129) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3586)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 130) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3587)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 131) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3588)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 132) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3589)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 133) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3590)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 134) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3591)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 135) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 4032)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129024)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4033)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129025)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4034)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129026)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4035)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129027)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4036)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129028)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4037)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129029)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4038)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129030)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4039)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129031)];
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4480)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 16)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4481)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 17)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4482)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 18)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4483)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 19)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4484)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 20)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4485)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 21)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4486)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 22)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4487)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 23)];
- }
- __syncthreads();
- for (int ry_outer_inner = 0; ry_outer_inner < 3; ++ry_outer_inner) {
- for (int rx_outer_inner = 0; rx_outer_inner < 3; ++rx_outer_inner) {
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 3)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 4)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 5)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 6)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 3)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 4)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 5)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 6)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 3)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 4)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 5)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 6)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 3)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 4)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 5)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 6)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 85)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 86)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 87)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 85)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 86)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 87)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 85)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 86)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 87)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 85)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 86)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 87)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 165)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 166)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 167)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 168)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 165)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 166)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 167)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 168)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 165)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 166)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 167)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 168)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 165)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 166)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 167)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 168)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 246)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 247)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 248)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 249)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 246)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 247)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 248)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 249)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 246)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 247)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 248)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 249)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 246)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 247)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 248)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 249)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 324)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 325)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 326)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 327)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 328)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 329)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 330)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 324)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 325)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 326)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 327)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 328)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 329)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 330)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 324)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 325)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 326)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 327)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 328)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 329)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 330)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 324)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 325)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 326)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 327)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 328)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 329)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 330)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 405)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 406)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 407)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 408)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 409)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 410)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 411)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 405)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 406)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 407)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 408)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 409)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 410)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 411)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 405)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 406)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 407)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 408)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 409)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 410)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 411)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 405)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 406)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 407)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 408)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 409)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 410)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 411)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 486)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 487)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 488)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 489)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 490)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 491)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 492)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 486)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 487)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 488)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 489)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 490)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 491)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 492)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 486)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 487)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 488)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 489)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 490)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 491)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 492)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 486)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 487)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 488)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 489)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 490)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 491)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 492)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 567)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 568)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 569)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 570)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 571)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 572)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 573)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 567)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 568)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 569)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 570)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 571)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 572)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 573)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 567)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 568)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 569)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 570)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 571)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 572)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 573)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 567)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 568)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 569)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 570)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 571)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 572)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 573)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 648)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 649)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 650)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 651)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 652)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 653)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 654)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 648)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 649)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 650)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 651)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 652)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 653)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 654)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 648)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 649)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 650)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 651)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 652)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 653)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 654)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 648)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 649)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 650)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 651)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 652)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 653)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 654)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 729)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 730)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 731)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 732)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 733)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 734)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 735)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 729)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 730)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 731)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 732)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 733)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 734)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 735)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 729)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 730)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 731)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 732)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 733)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 734)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 735)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 729)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 730)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 731)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 732)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 733)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 734)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 735)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 810)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 811)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 812)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 813)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 814)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 815)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 816)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 810)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 811)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 812)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 813)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 814)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 815)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 816)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 810)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 811)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 812)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 813)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 814)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 815)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 816)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 810)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 811)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 812)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 813)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 814)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 815)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 816)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 891)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 892)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 893)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 894)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 895)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 896)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 897)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 891)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 892)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 893)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 894)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 895)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 896)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 897)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 891)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 892)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 893)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 894)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 895)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 896)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 897)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 891)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 892)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 893)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 894)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 895)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 896)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 897)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 972)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 973)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 974)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 975)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 976)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 977)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 978)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 972)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 973)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 974)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 975)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 976)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 977)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 978)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 972)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 973)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 974)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 975)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 976)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 977)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 978)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 972)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 973)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 974)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 975)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 976)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 977)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 978)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1053)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1054)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1055)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1056)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1057)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1058)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1059)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1053)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1054)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1055)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1056)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1057)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1058)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1059)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1053)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1054)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1055)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1056)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1057)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1058)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1059)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1053)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1054)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1055)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1056)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1057)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1058)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1059)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1134)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1138)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1139)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1140)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1134)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1138)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1139)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1140)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1134)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1138)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1139)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1140)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1134)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1138)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1139)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1140)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1215)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1219)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1220)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1221)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1215)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1219)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1220)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1221)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1215)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1219)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1220)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1221)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1215)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1219)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1220)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1221)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
+ 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) {
+ for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((((int)threadIdx.x) % 49) / 7) + ry_outer_outer)) && ((((((int)threadIdx.x) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 2) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 2) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 104)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 4) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 4) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 216)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 336)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 6) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 6) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 328)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 1) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 1) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 440)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 560)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 3) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 3) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 552)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 672)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 5) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 5) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 664)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((1 <= (((((int)threadIdx.x) % 49) / 7) + ry_outer_outer)) && ((((((int)threadIdx.x) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 776)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 896)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 2) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 2) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 888)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1008)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 4) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 4) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1000)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1120)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 6) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 6) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1112)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1232)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 1) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 1) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1224)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1344)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 3) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 3) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1336)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1456)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 5) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 5) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1448)] : 0.000000e+00f);
+ kernel_shared[(((int)threadIdx.x) * 3)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) * 3) >> 5) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) * 3) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[((((int)threadIdx.x) * 3) + 1)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 1) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 1) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[((((int)threadIdx.x) * 3) + 2)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 2) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 2) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ if (((int)threadIdx.x) < 59) {
+ kernel_shared[((((int)threadIdx.x) * 3) + 336)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 336) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 16) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ }
+ if (((int)threadIdx.x) < 59) {
+ kernel_shared[((((int)threadIdx.x) * 3) + 337)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 337) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 17) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ }
+ if (((int)threadIdx.x) < 58) {
+ kernel_shared[((((int)threadIdx.x) * 3) + 338)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 338) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 18) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ }
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 16; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7))] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 49)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 50)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 51)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 52)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 53)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 54)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 55)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ }
}
}
}
- for (int i1_inner = 0; i1_inner < 4; ++i1_inner) {
- compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 1)] = max((conv2d_nchw[(i1_inner + 4)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 2)] = max((conv2d_nchw[(i1_inner + 8)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 3)] = max((conv2d_nchw[(i1_inner + 12)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 4)] = max((conv2d_nchw[(i1_inner + 16)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 5)] = max((conv2d_nchw[(i1_inner + 20)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 6)] = max((conv2d_nchw[(i1_inner + 24)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
+ for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
+ compute[(((((int)blockIdx.x) * 784) + (((int)threadIdx.x) * 7)) + i3_inner)] = max((conv2d_nchw[i3_inner] + bias[((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
}
}
@@ -1764,7 +616,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 6 minutes 25.850 seconds)
+ **Total running time of the script:** ( 6 minutes 28.831 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 0209933eb9..e81f52e392 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
@@ -647,7 +647,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 8.1379 8.1404 8.1508 8.1224 0.0117
+ 8.1199 8.1161 8.1291 8.1144 0.0065
@@ -675,7 +675,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 10.653 seconds)
+ **Total running time of the script:** ( 1 minutes 13.908 seconds)
.. _sphx_glr_download_how_to_tune_with_autoscheduler_tune_network_cuda.py:
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
index 68ff9c6e07..01e1769509 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
@@ -666,7 +666,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 757.0472 756.6426 758.7837 755.7152 1.2850
+ 770.8157 771.0289 771.6153 769.8030 0.7551
@@ -694,7 +694,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 42.813 seconds)
+ **Total running time of the script:** ( 1 minutes 47.644 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 9824948bb2..898fb7f092 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
@@ -389,27 +389,25 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
@T.prim_func
def main(placeholder: T.Buffer((128, 256), "float32"), placeholder_1: T.Buffer((4916, 16, 1), "float32"), placeholder_2: T.Buffer((4916,), "int32"), placeholder_3: T.Buffer((33,), "int32"), placeholder_4: T.Buffer((128, 512), "float32"), compute: T.Buffer((128, 512), "float32")):
T.func_attr({"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True})
- for i0_outer_i1_outer_fused in T.parallel(512):
- compute_1 = T.allocate([128], "float32", "global")
- compute_2 = T.Buffer((128,), data=compute_1)
- for i_outer_inner in range(2):
+ for i0_outer_i1_outer_fused in T.parallel(32):
+ compute_1 = T.allocate([2048], "float32", "global")
+ compute_2 = T.Buffer((2048,), data=compute_1)
+ for i_outer_inner in range(32):
for i_inner_init, j_init in T.grid(4, 16):
compute_2[i_outer_inner * 64 + i_inner_init * 16 + j_init] = T.float32(0)
- for elem_idx, i_inner, j in T.grid(T.Let(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1], where={cse_var_1: i0_outer_i1_outer_fused % 32}), 4, 16):
- cse_var_1 = T.int32()
+ for elem_idx, i_inner, j in T.grid(placeholder_5[i0_outer_i1_outer_fused + 1] - placeholder_5[i0_outer_i1_outer_fused], 4, 16):
placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_2: T.int32 = i0_outer_i1_outer_fused % 32
- if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ if T.likely(elem_idx < placeholder_5[i0_outer_i1_outer_fused + 1] - placeholder_5[i0_outer_i1_outer_fused]):
placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
placeholder_7 = T.Buffer((32768,), data=placeholder.data)
placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
- cse_var_3: T.int32 = i_outer_inner * 64 + i_inner * 16 + j
- compute_2[cse_var_3] = compute_2[cse_var_3] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 2048 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
- for i0_inner in range(8):
- cse_var_4: T.int32 = i0_outer_i1_outer_fused // 32 * 4096 + i0_inner * 512 + i0_outer_i1_outer_fused % 32 * 16
+ cse_var_1: T.int32 = i_outer_inner * 64 + i_inner * 16 + j
+ compute_2[cse_var_1] = compute_2[cse_var_1] + placeholder_6[placeholder_5[i0_outer_i1_outer_fused] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[i0_outer_i1_outer_fused] + elem_idx]], T.float32(0))
+ for i0_inner in range(128):
+ cse_var_2: T.int32 = i0_inner * 512 + i0_outer_i1_outer_fused * 16
compute_3 = T.Buffer((65536,), data=compute.data)
placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
- compute_3[cse_var_4:cse_var_4 + 16] = T.max(compute_2[i0_inner * 16:i0_inner * 16 + 16] + placeholder_5[cse_var_4:cse_var_4 + 16], T.Broadcast(T.float32(0), 16))
+ compute_3[cse_var_2:cse_var_2 + 16] = T.max(compute_2[i0_inner * 16:i0_inner * 16 + 16] + placeholder_5[cse_var_2:cse_var_2 + 16], T.Broadcast(T.float32(0), 16))
@@ -459,7 +457,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 1.361 ms
+ Execution time of this operator: 1.485 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 a937b61cde..e185c2fdd9 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
=================
-**00:37.756** total execution time for **how_to_tune_with_autotvm** files:
+**00:40.527** 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``) | 00:37.722 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:40.491 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.020 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.022 | 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``) | 00:00.005 | 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 |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``) | 00:00.005 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
index f7398e74fc..d35869dde7 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
@@ -390,8 +390,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 16, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10353238
- No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5478804
+ No: 2 GFLOPS: 114.52/114.52 result: MeasureResult(costs=(0.0020214115454545455,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.826885461807251, timestamp=1679626739.1992018) [('tile_f', [-1, 1, 1, 2]), ('tile_y', [-1, 1, 7, 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', 1)],None,6269615
+ No: 3 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -513,8 +514,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7754643
- No: 3 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4664196
+ No: 4 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -636,8 +637,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 64]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1050261
- No: 4 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 512, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4573854
+ No: 5 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -759,9 +760,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8029103
- No: 5 GFLOPS: 43.22/43.22 result: MeasureResult(costs=(0.005356821111111111,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6050300598144531, timestamp=1679616307.8310463) [('tile_f', [-1, 2, 64, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6649375
- No: 6 GFLOPS: 0.00/43.22 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 64, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4178335
+ No: 6 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -883,8 +883,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6375491
- No: 7 GFLOPS: 0.00/43.22 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7328010
+ No: 7 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1006,9 +1006,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8450931
- No: 8 GFLOPS: 0.99/43.22 result: MeasureResult(costs=(0.23476428200000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.164633512496948, timestamp=1679616311.447056) [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5811496
- No: 9 GFLOPS: 0.00/43.22 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9260858
+ No: 8 GFLOPS: 2.80/114.52 result: MeasureResult(costs=(0.08256726824999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.456061601638794, timestamp=1679626752.5319996) [('tile_f', [-1, 64, 2, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5821050
+ No: 9 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1130,8 +1130,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9463390
- No: 10 GFLOPS: 0.00/43.22 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2606293
+ No: 10 GFLOPS: 155.35/155.35 result: MeasureResult(costs=(0.0014902013766233768,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.0830042362213135, timestamp=1679626755.7926762) [('tile_f', [-1, 8, 4, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4879015
+ No: 11 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1253,10 +1254,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,107529
- No: 11 GFLOPS: 201.95/201.95 result: MeasureResult(costs=(0.0011463299279279278,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8168115615844727, timestamp=1679616315.2119982) [('tile_f', [-1, 2, 8, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,6988108
- No: 12 GFLOPS: 2.33/201.95 result: MeasureResult(costs=(0.0994267805,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.338953971862793, timestamp=1679616317.0621138) [('tile_f', [-1, 1, 1, 256]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2328916
- No: 13 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6578462
+ No: 12 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1378,8 +1377,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 8, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4705337
- No: 14 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5055302
+ No: 13 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1501,8 +1500,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9666307
- No: 15 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 8, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,342501
+ No: 14 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1624,8 +1623,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7221425
- No: 16 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5487563
+ No: 15 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1747,8 +1746,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 256, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,901832
- No: 17 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1293394
+ No: 16 GFLOPS: 259.48/259.48 result: MeasureResult(costs=(0.0008921655081967214,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.353220224380493, timestamp=1679626758.422771) [('tile_f', [-1, 1, 32, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9729540
+ No: 17 GFLOPS: 0.00/259.48 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1870,9 +1870,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 256, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5570672
- No: 18 GFLOPS: 53.19/201.95 result: MeasureResult(costs=(0.0043525985,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2984583377838135, timestamp=1679616319.7499006) [('tile_f', [-1, 1, 1, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7065656
- No: 19 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9016650
+ No: 18 GFLOPS: 0.00/259.48 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1994,8 +1993,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 1, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6032098
- No: 20 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 1, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,53302
+ No: 19 GFLOPS: 0.00/259.48 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2117,7 +2116,130 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 64]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7314326
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 1, 128]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,82932
+ No: 20 GFLOPS: 0.00/259.48 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
+ func = build(s, args, target=target, runtime=runtime)
+ File "/workspace/python/tvm/driver/build_module.py", line 227, in build
+ input_mod = lower(inputs, args, name=name, binds=binds)
+ File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
+ return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
+ File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+ tvm._ffi.base.TVMError: Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc:100
+ 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+ at ../include/tvm/runtime/packed_func.h:1753
+ 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+ at ../include/tvm/runtime/packed_func.h:1697
+ 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ at ../include/tvm/runtime/packed_func.h:1621
+ 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 1: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 0: operator()
+ at ../src/runtime/c_runtime_api.cc:534
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+
+ Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc:100
+ 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+ at ../include/tvm/runtime/packed_func.h:1753
+ 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+ at ../include/tvm/runtime/packed_func.h:1697
+ 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ at ../include/tvm/runtime/packed_func.h:1621
+ 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 1: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 0: operator()
+ at ../src/runtime/c_runtime_api.cc:534
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9460496
@@ -2172,9 +2294,9 @@ and measure running time.
Finish loading 20 records
Best config:
- [('tile_f', [-1, 2, 8, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,6988108
+ [('tile_f', [-1, 1, 32, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9729540
Finish loading 20 records
- Time cost of this operator: 0.001576
+ Time cost of this operator: 0.001312
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 945f098106..5aec556c6f 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
@@ -360,10 +360,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 315.5 98.744 (1, 2, 10, 10, 3) 2 1 [315.5]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.04 0.951 (1, 6, 10, 10) 1 1 [3.04]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.972 0.304 (1, 1, 10, 10, 3) 1 1 [0.972]
- Total_time - 319.512 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 314.2 98.697 (1, 2, 10, 10, 3) 2 1 [314.2]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.151 0.99 (1, 6, 10, 10) 1 1 [3.151]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.996 0.313 (1, 1, 10, 10, 3) 1 1 [0.996]
+ Total_time - 318.347 - - - - -
@@ -428,10 +428,10 @@ Timing the tuned program
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 100.2 97.31 (1, 6, 10, 10, 1) 2 1 [100.2]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.801 1.749 (1, 6, 10, 10) 1 1 [1.801]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.968 0.94 (1, 1, 10, 10, 3) 1 1 [0.968]
- Total_time - 102.97 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 136.8 98.078 (1, 6, 10, 10, 1) 2 1 [136.8]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.821 1.305 (1, 6, 10, 10) 1 1 [1.821]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.86 0.617 (1, 3, 10, 10, 1) 1 1 [0.86]
+ Total_time - 139.481 - - - - -
@@ -439,7 +439,7 @@ Timing the tuned program
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 20.905 seconds)
+ **Total running time of the script:** ( 1 minutes 28.427 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_autotune.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
index 9ba7b8adea..b0837484da 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
@@ -118,7 +118,7 @@ download a cat image and preprocess it to use as the model input.
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/ao/quantization/utils.py:281: UserWarning: must run observer before calling calculate_qparams. Returning default values.
"must run observer before calling calculate_qparams. " +
Downloading: "https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
-
0%| | 0.00/3.42M [00:00<?, ?B/s]
61%|###### | 2.09M/3.42M [00:00<00:00, 17.9MB/s]
100%|##########| 3.42M/3.42M [00:00<00:00, 27.9MB/s]
+
0%| | 0.00/3.42M [00:00<?, ?B/s]
61%|###### | 2.09M/3.42M [00:00<00:00, 16.8MB/s]
100%|##########| 3.42M/3.42M [00:00<00:00, 26.4MB/s]
/workspace/python/tvm/relay/frontend/pytorch_utils.py:47: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
return LooseVersion(torch_ver) > ver
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
@@ -324,7 +324,7 @@ Look up prediction top 1 index in 1000 class synset.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 16.988 seconds)
+ **Total running time of the script:** ( 1 minutes 22.970 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_pytorch.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
index 0732abf9a7..69cb8c3463 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
@@ -217,7 +217,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
.. code-block:: none
- '/tmp/tmpsq8t5s21/images/random'
+ '/tmp/tmpr8om6t_s/images/random'
@@ -308,7 +308,7 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. image-sg:: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
- :alt: [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]
+ :alt: [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0]
:srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
:class: sphx-glr-single-img
@@ -317,8 +317,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmpsq8t5s21/images/target contains 8144 images
- /tmp/tmpsq8t5s21/images/random contains 5000 images
+ /tmp/tmpr8om6t_s/images/target contains 8144 images
+ /tmp/tmpr8om6t_s/images/random contains 5000 images
@@ -493,13 +493,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 41s - loss: 0.2322 - accuracy: 0.9176 - val_loss: 0.1163 - val_accuracy: 0.9558 - 41s/epoch - 124ms/step
+ 328/328 - 42s - loss: 0.2768 - accuracy: 0.9157 - val_loss: 0.1518 - val_accuracy: 0.9517 - 42s/epoch - 128ms/step
Epoch 2/3
- 328/328 - 35s - loss: 0.0927 - accuracy: 0.9660 - val_loss: 0.0900 - val_accuracy: 0.9656 - 35s/epoch - 106ms/step
+ 328/328 - 35s - loss: 0.1149 - accuracy: 0.9561 - val_loss: 0.1394 - val_accuracy: 0.9513 - 35s/epoch - 108ms/step
Epoch 3/3
- 328/328 - 35s - loss: 0.0644 - accuracy: 0.9756 - val_loss: 0.1403 - val_accuracy: 0.9468 - 35s/epoch - 106ms/step
+ 328/328 - 35s - loss: 0.0679 - accuracy: 0.9748 - val_loss: 0.1042 - val_accuracy: 0.9653 - 35s/epoch - 107ms/step
- <keras.callbacks.History object at 0x7fc54d3a6890>
+ <keras.callbacks.History object at 0x7f228f798990>
@@ -860,7 +860,7 @@ Arduino tutorial for how to do that `on GitHub <https://github.com/guberti/tvm-a
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 4 minutes 22.921 seconds)
+ **Total running time of the script:** ( 4 minutes 31.483 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 bc50b1e03b..0fcd9313a0 100644
--- a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
@@ -5,20 +5,20 @@
Computation times
=================
-**07:26.229** total execution time for **how_to_work_with_microtvm** files:
+**07:50.419** total execution time for **how_to_work_with_microtvm** files:
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:22.921 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:31.483 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 01:20.905 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 01:28.427 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``) | 01:16.988 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``) | 01:22.970 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:10.223 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:10.957 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:07.873 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:08.651 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:07.318 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:07.931 | 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 7a4ed8af3b..30d0e8430c 100644
--- a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
@@ -5,14 +5,14 @@
Computation times
=================
-**00:36.852** total execution time for **how_to_work_with_relay** files:
+**00:38.525** total execution time for **how_to_work_with_relay** files:
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:32.206 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:33.774 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:02.895 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:03.004 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.744 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.740 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``) | 00:00.007 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
index 10a025c3e1..bd50c0fa5b 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
@@ -264,7 +264,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
.. code-block:: none
- <function my_cuda_math_rule at 0x7fc19cc05d40>
+ <function my_cuda_math_rule at 0x7f1edd6a7710>
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 26c0469b69..8c77395aaa 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
=================
-**00:07.653** total execution time for **how_to_work_with_schedules** files:
+**00:08.721** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:04.907 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:05.745 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.241 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.410 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.620 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.642 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.604 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.631 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.133 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.138 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.062 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.066 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.056 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.058 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.030 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.032 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index 56219c46a9..fc2cb8517c 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
=================
-**00:30.634** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:33.229** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:30.628 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:33.222 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``) | 00:00.006 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``) | 00:00.007 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
index cf903ae935..b5ea67b875 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -293,7 +293,7 @@ The compilation steps are:
DeprecationWarning,
/workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the new recommended usage.
relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
- resnet18_v1 inference graph built in 32.81s!
+ resnet18_v1 inference graph built in 36.54s!
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 dccd43ce84..0e5be90dc4 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -337,7 +337,7 @@ The compilation steps are:
/workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 22.30s!
+ yolov3-tiny inference graph built in 24.42s!
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 d5dca69fac..036defa449 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
=================
-**01:39.454** total execution time for **topic_vta_tutorials_frontend** files:
+**01:45.411** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:49.770 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:53.678 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:49.683 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:51.733 | 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 f0c36d52da..32220e4956 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
=================
-**00:03.153** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.242** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.672 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.729 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.481 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.513 | 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 b3e2d6b31a..3121de42d5 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
=================
-**00:00.814** total execution time for **topic_vta_tutorials** files:
+**00:00.884** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.420 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.457 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.394 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.427 | 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 cbfac0662c..2213e38b76 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -318,7 +318,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 96.193 ms
+ Execution time of this operator: 100.398 ms
@@ -434,7 +434,7 @@ operations.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 32.083 seconds)
+ **Total running time of the script:** ( 1 minutes 24.282 seconds)
.. _sphx_glr_download_tutorial_auto_scheduler_matmul_x86.py:
diff --git a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
index 552feee64c..2598c8fed7 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -454,16 +454,16 @@ reduce variance, we take 5 measurements and average them.
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 8.86/8.86 result: MeasureResult(costs=(0.030295845000000005,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.111386775970459, timestamp=1679614657.2000778) [('tile_y', [-1, 16]), ('tile_x', [-1, 128])],None,74
- No: 2 GFLOPS: 10.48/10.48 result: MeasureResult(costs=(0.0256183964,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6739628314971924, timestamp=1679614657.8733046) [('tile_y', [-1, 4]), ('tile_x', [-1, 32])],None,52
- No: 3 GFLOPS: 10.65/10.65 result: MeasureResult(costs=(0.0252048924,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7194416522979736, timestamp=1679614659.845875) [('tile_y', [-1, 4]), ('tile_x', [-1, 128])],None,72
- No: 4 GFLOPS: 0.95/10.65 result: MeasureResult(costs=(0.2813616966,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.752430438995361, timestamp=1679614664.6092327) [('tile_y', [-1, 32]), ('tile_x', [-1, 2])],None,15
- No: 5 GFLOPS: 1.98/10.65 result: MeasureResult(costs=(0.13565037119999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.4148261547088623, timestamp=1679614667.168229) [('tile_y', [-1, 8]), ('tile_x', [-1, 1])],None,3
- No: 6 GFLOPS: 9.84/10.65 result: MeasureResult(costs=(0.027278074199999995,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6956429481506348, timestamp=1679614669.171133) [('tile_y', [-1, 2]), ('tile_x', [-1, 256])],None,81
- No: 7 GFLOPS: 11.26/11.26 result: MeasureResult(costs=(0.023835292799999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6453514099121094, timestamp=1679614671.1168091) [('tile_y', [-1, 128]), ('tile_x', [-1, 32])],None,57
- No: 8 GFLOPS: 3.39/11.26 result: MeasureResult(costs=(0.07919194,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5253140926361084, timestamp=1679614672.6409886) [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
- No: 9 GFLOPS: 4.39/11.26 result: MeasureResult(costs=(0.061089533200000004,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2082717418670654, timestamp=1679614674.0398715) [('tile_y', [-1, 16]), ('tile_x', [-1, 16])],None,44
- No: 10 GFLOPS: 12.52/12.52 result: MeasureResult(costs=(0.021447108,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6481060981750488, timestamp=1679614674.6472747) [('tile_y', [-1, 256]), ('tile_x', [-1, 128])],None,78
+ No: 1 GFLOPS: 10.03/10.03 result: MeasureResult(costs=(0.026752570600000004,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7407488822937012, timestamp=1679625052.3152878) [('tile_y', [-1, 16]), ('tile_x', [-1, 32])],None,54
+ No: 2 GFLOPS: 11.13/11.13 result: MeasureResult(costs=(0.0241221214,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.649803638458252, timestamp=1679625052.971663) [('tile_y', [-1, 128]), ('tile_x', [-1, 512])],None,97
+ No: 3 GFLOPS: 11.54/11.54 result: MeasureResult(costs=(0.0232565404,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6532163619995117, timestamp=1679625054.9199133) [('tile_y', [-1, 8]), ('tile_x', [-1, 256])],None,83
+ No: 4 GFLOPS: 8.93/11.54 result: MeasureResult(costs=(0.0300584328,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7522788047790527, timestamp=1679625056.9847775) [('tile_y', [-1, 256]), ('tile_x', [-1, 256])],None,88
+ No: 5 GFLOPS: 12.50/12.50 result: MeasureResult(costs=(0.02148152,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6188480854034424, timestamp=1679625057.795037) [('tile_y', [-1, 32]), ('tile_x', [-1, 128])],None,75
+ No: 6 GFLOPS: 7.39/12.50 result: MeasureResult(costs=(0.0363478626,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.9304215908050537, timestamp=1679625059.9495907) [('tile_y', [-1, 1]), ('tile_x', [-1, 64])],None,60
+ No: 7 GFLOPS: 0.51/12.50 result: MeasureResult(costs=(0.5292683636,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.71769094467163, timestamp=1679625068.6794431) [('tile_y', [-1, 128]), ('tile_x', [-1, 1])],None,7
+ No: 8 GFLOPS: 2.10/12.50 result: MeasureResult(costs=(0.1277711432,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.311182975769043, timestamp=1679625070.9895515) [('tile_y', [-1, 128]), ('tile_x', [-1, 4])],None,27
+ No: 9 GFLOPS: 10.56/12.50 result: MeasureResult(costs=(0.025417857,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6243619918823242, timestamp=1679625071.7493625) [('tile_y', [-1, 2]), ('tile_x', [-1, 512])],None,91
+ No: 10 GFLOPS: 11.04/12.50 result: MeasureResult(costs=(0.024324382999999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6199319362640381, timestamp=1679625072.4000835) [('tile_y', [-1, 128]), ('tile_x', [-1, 32])],None,57
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index e6d058a7b4..b8c7086e77 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -311,7 +311,7 @@ standard deviation.
.. code-block:: none
- {'mean': 513.497117569998, 'median': 513.9784247000023, 'std': 2.379501423445631}
+ {'mean': 512.8374819899979, 'median': 513.379893900003, 'std': 2.048731521479577}
@@ -545,32 +545,29 @@ the tuning data to.
.. code-block:: none
-
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 6.77/ 15.66 GFLOPS | Progress: (4/20) | 12.27 s
[Task 1/25] Current/Best: 18.37/ 18.37 GFLOPS | Progress: (8/20) | 15.88 s
[Task 1/25] Current/Best: 22.95/ 23.24 GFLOPS | Progress: (12/20) | 19.64 s
[Task 1/25] Current/Best: 11.16/ 23.24 GFLOPS | Progress: (16/20) | 22.56 s
[Task 1/25] Current/Best: 3.16/ 23.24 GFLOPS | Progress: (20/20) | 25.84 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 19.04/ 19.04 GFLOPS | Progress: (4/20) | 4.43 s
[Task 2/25] Current/Best: 10.58/ 20.32 GFLOPS | Progress: (8/20) | 7.51 s
[Task 2/25] Current/Best: 6.76/ 20.32 GFLOPS | Progress: (12/20) | 9.00 s
[Task 2/25] Current/Best: 21.01/ 21.01 GFLOPS | Progress: (16/20) | 10.75 s
[Task 2/25] Current/Best: 12.42/ 22.67 GFLOPS | Progress: (20/20) | 12.47 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 7.55/ 18.91 GFLOPS | Progress: (4/20) | 5.32 s
[Task 3/25] Current/Best: 8.17/ 18.91 GFLOPS | Progress: (8/20) | 7.74 s
[Task 3/25] Current/Best: 6.53/ 18.91 GFLOPS | Progress: (12/20) | 10.25 s
[Task 3/25] Current/Best: 17.18/ 18.91 GFLOPS | Progress: (16/20) | 12.68 s
[Task 3/25] Current/Best: 17.26/ 24.16 GFLOPS | Progress: (20/20) | 14.97 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 21.56/ 21.56 GFLOPS | Progress: (4/20) | 4.50 s
[Task 4/25] Current/Best: 11.26/ 21.56 GFLOPS | Progress: (8/20) | 9.71 s
[Task 4/25] Current/Best: 13.22/ 21.56 GFLOPS | Progress: (12/20) | 11.96 s
[Task 4/25] Current/Best: 12.55/ 21.56 GFLOPS | Progress: (16/20) | 14.03 s
[Task 4/25] Current/Best: 9.06/ 21.56 GFLOPS | Progress: (20/20) | 18.93 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 18.31/ 18.31 GFLOPS | Progress: (4/20) | 5.32 s
[Task 5/25] Current/Best: 3.73/ 18.31 GFLOPS | Progress: (8/20) | 7.92 s
[Task 5/25] Current/Best: 5.42/ 18.31 GFLOPS | Progress: (12/20) | 9.97 s
[Task 5/25] Current/Best: 6.76/ 20.84 GFLOPS | Progress: (16/20) | 11.87 s
[Task 5/25] Current/Best: 16.05/ 20.84 GFLOPS | Progress: (20/20) | 13.73 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 11.51/ 15.94 GFLOPS | Progress: (4/20) | 5.42 s
[Task 6/25] Current/Best: 9.69/ 15.94 GFLOPS | Progress: (8/20) | 8.15 s
[Task 6/25] Current/Best: 3.03/ 15.94 GFLOPS | Progress: (12/20) | 11.60 s
[Task 6/25] Current/Best: 17.08/ 18.22 GFLOPS | Progress: (16/20) | 14.89 s
[Task 6/25] Current/Best: 4.98/ 18.22 GFLOPS | Progress: (20/20) | 18.18 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 23.15/ 23.15 GFLOPS | Progress: (4/20) | 5.20 s
[Task 7/25] Current/Best: 8.72/ 23.15 GFLOPS | Progress: (8/20) | 7.58 s
[Task 7/25] Current/Best: 13.40/ 23.15 GFLOPS | Progress: (12/20) | 9.72 s
[Task 7/25] Current/Best: 12.27/ 23.15 GFLOPS | Progress: (16/20) | 11.88 s
[Task 7/25] Current/Best: 7.74/ 23.15 GFLOPS | Progress: (20/20) | 14.46 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 2.09/ 19.32 GFLOPS | Progress: (4/20) | 6.59 s
[Task 8/25] Current/Best: 13.87/ 19.32 GFLOPS | Progress: (8/20) | 10.45 s
[Task 8/25] Current/Best: 10.84/ 19.32 GFLOPS | Progress: (12/20) | 14.48 s
[Task 8/25] Current/Best: 10.00/ 19.32 GFLOPS | Progress: (16/20) | 19.76 s
[Task 8/25] Current/Best: 14.10/ 19.32 GFLOPS | Progress: (20/20) | 25.26 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 11.80/ 20.04 GFLOPS | Progress: (4/20) | 4.97 s
[Task 9/25] Current/Best: 6.75/ 20.04 GFLOPS | Progress: (8/20) | 6.77 s
[Task 9/25] Current/Best: 11.97/ 20.04 GFLOPS | Progress: (12/20) | 10.39 s
[Task 9/25] Current/Best: 7.80/ 20.04 GFLOPS | Progress: (16/20) | 19.32 s
[Task 9/25] Current/Best: 11.39/ 20.04 GFLOPS | Progress: (20/20) | 22.64 s Done.
-
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 13.57/ 13.57 GFLOPS | Progress: (4/20) | 4.91 s
[Task 10/25] Current/Best: 12.34/ 14.42 GFLOPS | Progress: (8/20) | 7.12 s
[Task 10/25] Current/Best: 14.18/ 15.09 GFLOPS | Progress: (12/20) | 9.35 s
[Task 10/25] Current/Best: 12.34/ 15.09 GFLOPS | Progress: (16/20) | 11.97 s
[Task 10/25] Current/Best: 14.22/ 15.53 GFLOPS | Progress: (20/20) | 13.80 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 20.51/ 20.51 GFLOPS | Progress: (4/20) | 4.85 s
[Task 11/25] Current/Best: 16.11/ 20.51 GFLOPS | Progress: (8/20) | 9.26 s
[Task 11/25] Current/Best: 11.36/ 20.51 GFLOPS | Progress: (12/20) | 12.26 s
[Task 11/25] Current/Best: 16.43/ 21.61 GFLOPS | Progress: (16/20) | 14.81 s
[Task 11/25] Current/Best: 12.32/ 21.61 GFLOPS | Progress: (20/20) | 17.14 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 20.24/ 20.24 GFLOPS | Progress: (4/20) | 5.59 s
[Task 12/25] Current/Best: 2.95/ 20.24 GFLOPS | Progress: (8/20) | 8.27 s
[Task 12/25] Current/Best: 11.39/ 20.24 GFLOPS | Progress: (12/20) | 15.21 s
[Task 12/25] Current/Best: 12.59/ 20.24 GFLOPS | Progress: (16/20) | 21.11 s
[Task 12/25] Current/Best: 11.14/ 20.24 GFLOPS | Progress: (20/20) | 24.57 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 3.09/ 20.32 GFLOPS | Progress: (4/20) | 5.69 s
[Task 13/25] Current/Best: 6.15/ 20.80 GFLOPS | Progress: (8/20) | 8.63 s
[Task 13/25] Current/Best: 13.03/ 20.80 GFLOPS | Progress: (12/20) | 10.76 s
[Task 13/25] Current/Best: 8.67/ 20.80 GFLOPS | Progress: (16/20) | 16.19 s
[Task 13/25] Current/Best: 9.65/ 23.18 GFLOPS | Progress: (20/20) | 18.98 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 12.97/ 17.93 GFLOPS | Progress: (4/20) | 5.44 s
[Task 14/25] Current/Best: 14.68/ 18.12 GFLOPS | Progress: (8/20) | 7.84 s
[Task 14/25] Current/Best: 11.49/ 18.12 GFLOPS | Progress: (12/20) | 19.22 s
[Task 14/25] Current/Best: 10.78/ 18.12 GFLOPS | Progress: (16/20) | 21.50 s
[Task 14/25] Current/Best: 14.34/ 18.12 GFLOPS | Progress: (20/20) | 25.78 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
[Task 15/25] Current/Best: 15.71/ 15.71 GFLOPS | Progress: (4/20) | 9.10 s
[Task 15/25] Current/Best: 12.40/ 15.71 GFLOPS | Progress: (8/20) | 10.97 s
[Task 15/25] Current/Best: 13.34/ 20.85 GFLOPS | Progress: (12/20) | 13.73 s
[Task 15/25] Current/Best: 9.50/ 20.85 GFLOPS | Progress: (16/20) | 15.98 s
[Task 15/25] Current/Best: 15.97/ 20.85 GFLOPS | Progress: (20/20) | 17.97 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 16.75/ 20.44 GFLOPS | Progress: (4/20) | 5.65 s
[Task 16/25] Current/Best: 19.82/ 20.44 GFLOPS | Progress: (8/20) | 7.26 s
[Task 16/25] Current/Best: 14.66/ 20.44 GFLOPS | Progress: (12/20) | 9.03 s
[Task 16/25] Current/Best: 15.97/ 20.44 GFLOPS | Progress: (16/20) | 10.72 s
[Task 16/25] Current/Best: 15.28/ 20.44 GFLOPS | Progress: (20/20) | 12.54 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 14.23/ 14.23 GFLOPS | Progress: (4/20) | 5.48 s
[Task 17/25] Current/Best: 6.21/ 14.23 GFLOPS | Progress: (8/20) | 9.28 s
[Task 17/25] Current/Best: 22.30/ 22.30 GFLOPS | Progress: (12/20) | 11.46 s
[Task 17/25] Current/Best: 19.93/ 22.30 GFLOPS | Progress: (16/20) | 13.70 s
[Task 17/25] Current/Best: 20.53/ 22.30 GFLOPS | Progress: (20/20) | 15.61 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 8.22/ 21.50 GFLOPS | Progress: (4/20) | 4.79 s
[Task 18/25] Current/Best: 5.03/ 21.50 GFLOPS | Progress: (8/20) | 11.32 s
[Task 18/25] Current/Best: 11.11/ 21.50 GFLOPS | Progress: (12/20) | 13.40 s
[Task 18/25] Current/Best: 5.40/ 21.50 GFLOPS | Progress: (16/20) | 16.22 s
[Task 18/25] Current/Best: 13.78/ 21.50 GFLOPS | Progress: (20/20) | 22.17 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 6.16/ 11.25 GFLOPS | Progress: (4/20) | 7.67 s
[Task 19/25] Current/Best: 6.37/ 19.36 GFLOPS | Progress: (8/20) | 12.09 s
[Task 19/25] Current/Best: 19.30/ 19.36 GFLOPS | Progress: (12/20) | 15.22 s
[Task 19/25] Current/Best: 1.55/ 19.36 GFLOPS | Progress: (16/20) | 22.31 s
[Task 19/25] Current/Best: 20.54/ 20.54 GFLOPS | Progress: (20/20) | 25.74 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 6.17/ 12.70 GFLOPS | Progress: (4/20) | 5.12 s
[Task 20/25] Current/Best: 10.60/ 15.81 GFLOPS | Progress: (8/20) | 8.51 s
[Task 20/25] Current/Best: 9.70/ 19.41 GFLOPS | Progress: (12/20) | 16.58 s
[Task 20/25] Current/Best: 13.77/ 19.41 GFLOPS | Progress: (16/20) | 18.42 s Done.
-
[Task 20/25] Current/Best: 8.06/ 19.41 GFLOPS | Progress: (20/20) | 21.75 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 9.01/ 10.52 GFLOPS | Progress: (4/20) | 6.48 s
[Task 21/25] Current/Best: 18.77/ 18.77 GFLOPS | Progress: (8/20) | 11.16 s
[Task 21/25] Current/Best: 19.82/ 21.81 GFLOPS | Progress: (12/20) | 12.74 s
[Task 21/25] Current/Best: 17.79/ 21.81 GFLOPS | Progress: (16/20) | 15.67 s
[Task 21/25] Current/Best: 17.10/ 21.81 GFLOPS | Progress: (20/20) | 18.93 s Done.
-
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 9.94/ 9.94 GFLOPS | Progress: (4/20) | 5.92 s
[Task 22/25] Current/Best: 5.26/ 14.37 GFLOPS | Progress: (8/20) | 7.81 s
[Task 22/25] Current/Best: 11.33/ 17.79 GFLOPS | Progress: (12/20) | 9.71 s
[Task 22/25] Current/Best: 7.20/ 17.79 GFLOPS | Progress: (16/20) | 13.01 s
[Task 22/25] Current/Best: 20.11/ 20.11 GFLOPS | Progress: (20/20) | 15.72 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 23.52/ 23.52 GFLOPS | Progress: (4/20) | 5.30 s
[Task 23/25] Current/Best: 11.80/ 23.52 GFLOPS | Progress: (8/20) | 7.98 s
[Task 23/25] Current/Best: 20.69/ 23.52 GFLOPS | Progress: (12/20) | 13.35 s
[Task 23/25] Current/Best: 19.00/ 23.52 GFLOPS | Progress: (16/20) | 15.76 s
[Task 23/25] Current/Best: 12.07/ 23.52 GFLOPS | Progress: (20/20) | 19.75 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.34/ 8.34 GFLOPS | Progress: (4/20) | 13.47 s
[Task 24/25] Current/Best: 8.55/ 8.96 GFLOPS | Progress: (8/20) | 25.72 s
[Task 24/25] Current/Best: 3.43/ 8.96 GFLOPS | Progress: (12/20) | 31.53 s
[Task 24/25] Current/Best: 6.76/ 8.96 GFLOPS | Progress: (16/20) | 42.18 s
[Task 24/25] Current/Best: 3.44/ 8.96 GFLOPS | Progress: (20/20) | 54.54 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
[Task 25/25] Current/Best: 2.88/ 3.30 GFLOPS | Progress: (4/20) | 15.81 s
[Task 25/25] Current/Best: 9.49/ 9.49 GFLOPS | Progress: (8/20) | 22.07 s
[Task 25/25] Current/Best: 8.45/ 9.49 GFLOPS | Progress: (12/20) | 33.05 s
[Task 25/25] Current/Best: 8.77/ 9.49 GFLOPS | Progress: (16/20) | 37.29 s Done.
-
[Task 25/25] Current/Best: 3.03/ 9.49 GFLOPS | Progress: (20/20) | 45.86 s Done.
-
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 12.64/ 17.36 GFLOPS | Progress: (4/20) | 12.33 s
[Task 1/25] Current/Best: 15.05/ 17.36 GFLOPS | Progress: (8/20) | 16.51 s
[Task 1/25] Current/Best: 6.72/ 17.36 GFLOPS | Progress: (12/20) | 21.17 s
[Task 1/25] Current/Best: 8.38/ 17.36 GFLOPS | Progress: (16/20) | 23.68 s
[Task 1/25] Current/Best: 15.33/ 17.36 GFLOPS | Progress: (20/20) | 26.12 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.04/ 12.04 GFLOPS | Progress: (4/20) | 4.42 s
[Task 2/25] Current/Best: 13.62/ 18.02 GFLOPS | Progress: (8/20) | 6.62 s
[Task 2/25] Current/Best: 9.94/ 18.02 GFLOPS | Progress: (12/20) | 8.40 s
[Task 2/25] Current/Best: 17.13/ 18.02 GFLOPS | Progress: (16/20) | 10.38 s
[Task 2/25] Current/Best: 12.44/ 19.88 GFLOPS | Progress: (20/20) | 12.05 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 11.38/ 13.39 GFLOPS | Progress: (4/20) | 5.51 s
[Task 3/25] Current/Best: 20.78/ 20.78 GFLOPS | Progress: (8/20) | 7.93 s
[Task 3/25] Current/Best: 3.13/ 24.29 GFLOPS | Progress: (12/20) | 10.67 s
[Task 3/25] Current/Best: 10.26/ 24.29 GFLOPS | Progress: (16/20) | 12.98 s
[Task 3/25] Current/Best: 10.09/ 24.29 GFLOPS | Progress: (20/20) | 15.10 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 11.76/ 18.70 GFLOPS | Progress: (4/20) | 5.93 s
[Task 4/25] Current/Best: 8.78/ 18.80 GFLOPS | Progress: (8/20) | 9.05 s
[Task 4/25] Current/Best: 15.65/ 18.80 GFLOPS | Progress: (12/20) | 11.83 s
[Task 4/25] Current/Best: 11.77/ 18.80 GFLOPS | Progress: (16/20) | 14.86 s
[Task 4/25] Current/Best: 7.81/ 18.80 GFLOPS | Progress: (20/20) | 17.00 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 3.18/ 16.10 GFLOPS | Progress: (4/20) | 5.51 s
[Task 5/25] Current/Best: 10.22/ 16.10 GFLOPS | Progress: (8/20) | 8.05 s
[Task 5/25] Current/Best: 10.38/ 17.98 GFLOPS | Progress: (12/20) | 11.51 s
[Task 5/25] Current/Best: 12.33/ 19.56 GFLOPS | Progress: (16/20) | 13.61 s
[Task 5/25] Current/Best: 14.14/ 19.56 GFLOPS | Progress: (20/20) | 15.37 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 11.86/ 11.86 GFLOPS | Progress: (4/20) | 5.98 s
[Task 6/25] Current/Best: 17.89/ 17.89 GFLOPS | Progress: (8/20) | 9.24 s
[Task 6/25] Current/Best: 11.39/ 20.41 GFLOPS | Progress: (12/20) | 11.54 s
[Task 6/25] Current/Best: 15.14/ 20.41 GFLOPS | Progress: (16/20) | 15.87 s
[Task 6/25] Current/Best: 4.72/ 20.41 GFLOPS | Progress: (20/20) | 18.98 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 9.62/ 20.97 GFLOPS | Progress: (4/20) | 5.07 s
[Task 7/25] Current/Best: 15.02/ 20.97 GFLOPS | Progress: (8/20) | 7.08 s
[Task 7/25] Current/Best: 11.07/ 20.97 GFLOPS | Progress: (12/20) | 9.93 s
[Task 7/25] Current/Best: 17.47/ 20.97 GFLOPS | Progress: (16/20) | 12.62 s
[Task 7/25] Current/Best: 15.51/ 20.97 GFLOPS | Progress: (20/20) | 15.02 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 8.60/ 17.59 GFLOPS | Progress: (4/20) | 7.00 s
[Task 8/25] Current/Best: 8.21/ 17.59 GFLOPS | Progress: (8/20) | 11.96 s
[Task 8/25] Current/Best: 19.86/ 19.86 GFLOPS | Progress: (12/20) | 14.00 s
[Task 8/25] Current/Best: 12.89/ 19.86 GFLOPS | Progress: (16/20) | 18.13 s
[Task 8/25] Current/Best: 16.33/ 19.86 GFLOPS | Progress: (20/20) | 22.00 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 8.56/ 20.82 GFLOPS | Progress: (4/20) | 14.22 s
[Task 9/25] Current/Best: 9.08/ 20.82 GFLOPS | Progress: (8/20) | 17.92 s
[Task 9/25] Current/Best: 6.01/ 20.82 GFLOPS | Progress: (12/20) | 25.78 s
[Task 9/25] Current/Best: 21.81/ 21.81 GFLOPS | Progress: (16/20) | 29.01 s
[Task 9/25] Current/Best: 13.27/ 21.81 GFLOPS | Progress: (20/20) | 30.83 s Done.
+
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 6.10/ 20.34 GFLOPS | Progress: (4/20) | 4.79 s
[Task 10/25] Current/Best: 5.26/ 20.34 GFLOPS | Progress: (8/20) | 7.02 s
[Task 10/25] Current/Best: 12.36/ 20.34 GFLOPS | Progress: (12/20) | 9.50 s
[Task 10/25] Current/Best: 13.67/ 20.34 GFLOPS | Progress: (16/20) | 11.77 s
[Task 10/25] Current/Best: 9.19/ 20.34 GFLOPS | Progress: (20/20) | 14.96 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 17.38/ 20.84 GFLOPS | Progress: (4/20) | 5.40 s
[Task 11/25] Current/Best: 13.11/ 20.84 GFLOPS | Progress: (8/20) | 8.07 s
[Task 11/25] Current/Best: 10.83/ 20.84 GFLOPS | Progress: (12/20) | 10.75 s
[Task 11/25] Current/Best: 9.73/ 20.84 GFLOPS | Progress: (16/20) | 13.39 s
[Task 11/25] Current/Best: 7.13/ 20.84 GFLOPS | Progress: (20/20) | 15.96 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 11.21/ 13.36 GFLOPS | Progress: (4/20) | 6.10 s
[Task 12/25] Current/Best: 10.82/ 13.36 GFLOPS | Progress: (8/20) | 9.41 s
[Task 12/25] Current/Best: 14.81/ 18.80 GFLOPS | Progress: (12/20) | 12.50 s
[Task 12/25] Current/Best: 13.27/ 18.80 GFLOPS | Progress: (16/20) | 15.82 s
[Task 12/25] Current/Best: 12.89/ 18.80 GFLOPS | Progress: (20/20) | 20.23 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 5.17/ 17.55 GFLOPS | Progress: (4/20) | 5.72 s
[Task 13/25] Current/Best: 16.75/ 17.55 GFLOPS | Progress: (8/20) | 8.35 s
[Task 13/25] Current/Best: 12.11/ 17.55 GFLOPS | Progress: (12/20) | 12.48 s
[Task 13/25] Current/Best: 1.56/ 18.01 GFLOPS | Progress: (16/20) | 16.13 s
[Task 13/25] Current/Best: 16.27/ 20.63 GFLOPS | Progress: (20/20) | 18.72 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 6.03/ 18.11 GFLOPS | Progress: (4/20) | 5.48 s
[Task 14/25] Current/Best: 7.72/ 20.36 GFLOPS | Progress: (8/20) | 8.21 s
[Task 14/25] Current/Best: 17.75/ 20.36 GFLOPS | Progress: (12/20) | 10.66 s
[Task 14/25] Current/Best: 9.24/ 20.36 GFLOPS | Progress: (16/20) | 14.82 s
[Task 14/25] Current/Best: 13.29/ 20.36 GFLOPS | Progress: (20/20) | 20.06 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
[Task 15/25] Current/Best: 11.95/ 11.95 GFLOPS | Progress: (4/20) | 6.28 s
[Task 15/25] Current/Best: 20.45/ 20.45 GFLOPS | Progress: (8/20) | 9.17 s
[Task 15/25] Current/Best: 8.68/ 20.45 GFLOPS | Progress: (12/20) | 12.10 s
[Task 15/25] Current/Best: 9.93/ 20.45 GFLOPS | Progress: (16/20) | 14.95 s
[Task 15/25] Current/Best: 20.97/ 20.97 GFLOPS | Progress: (20/20) | 17.87 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 5.95/ 18.21 GFLOPS | Progress: (4/20) | 5.08 s
[Task 16/25] Current/Best: 16.04/ 21.43 GFLOPS | Progress: (8/20) | 7.12 s
[Task 16/25] Current/Best: 10.50/ 21.43 GFLOPS | Progress: (12/20) | 9.17 s
[Task 16/25] Current/Best: 20.66/ 21.43 GFLOPS | Progress: (16/20) | 12.79 s
[Task 16/25] Current/Best: 5.71/ 21.43 GFLOPS | Progress: (20/20) | 15.04 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 22.18/ 22.18 GFLOPS | Progress: (4/20) | 5.22 s
[Task 17/25] Current/Best: 10.24/ 22.18 GFLOPS | Progress: (8/20) | 7.53 s
[Task 17/25] Current/Best: 9.84/ 22.18 GFLOPS | Progress: (12/20) | 10.45 s
[Task 17/25] Current/Best: 6.20/ 22.67 GFLOPS | Progress: (16/20) | 12.77 s
[Task 17/25] Current/Best: 9.48/ 22.67 GFLOPS | Progress: (20/20) | 15.75 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 9.36/ 13.54 GFLOPS | Progress: (4/20) | 5.25 s
[Task 18/25] Current/Best: 15.16/ 17.87 GFLOPS | Progress: (8/20) | 7.93 s
[Task 18/25] Current/Best: 6.30/ 18.81 GFLOPS | Progress: (12/20) | 10.06 s
[Task 18/25] Current/Best: 6.68/ 19.80 GFLOPS | Progress: (16/20) | 12.11 s
[Task 18/25] Current/Best: 12.52/ 19.80 GFLOPS | Progress: (20/20) | 15.23 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 11.68/ 14.31 GFLOPS | Progress: (4/20) | 6.02 s
[Task 19/25] Current/Best: 3.08/ 21.42 GFLOPS | Progress: (8/20) | 9.81 s
[Task 19/25] Current/Best: 9.99/ 22.10 GFLOPS | Progress: (12/20) | 12.62 s
[Task 19/25] Current/Best: 3.08/ 22.10 GFLOPS | Progress: (16/20) | 16.86 s
[Task 19/25] Current/Best: 9.96/ 22.10 GFLOPS | Progress: (20/20) | 22.38 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 9.02/ 11.31 GFLOPS | Progress: (4/20) | 8.38 s
[Task 20/25] Current/Best: 17.26/ 17.26 GFLOPS | Progress: (8/20) | 11.16 s
[Task 20/25] Current/Best: 10.68/ 17.26 GFLOPS | Progress: (12/20) | 13.17 s
[Task 20/25] Current/Best: 10.73/ 17.26 GFLOPS | Progress: (16/20) | 16.87 s
[Task 20/25] Current/Best: 11.54/ 19.17 GFLOPS | Progress: (20/20) | 20.02 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 8.83/ 17.39 GFLOPS | Progress: (4/20) | 10.61 s
[Task 21/25] Current/Best: 15.93/ 22.25 GFLOPS | Progress: (8/20) | 12.39 s
[Task 21/25] Current/Best: 19.25/ 22.25 GFLOPS | Progress: (12/20) | 14.75 s
[Task 21/25] Current/Best: 5.21/ 22.25 GFLOPS | Progress: (16/20) | 20.06 s
[Task 21/25] Current/Best: 10.65/ 22.25 GFLOPS | Progress: (20
/20) | 23.05 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 12.40/ 17.15 GFLOPS | Progress: (4/20) | 4.95 s
[Task 22/25] Current/Best: 18.65/ 22.78 GFLOPS | Progress: (8/20) | 7.13 s
[Task 22/25] Current/Best: 4.72/ 22.78 GFLOPS | Progress: (12/20) | 9.31 s
[Task 22/25] Current/Best: 10.51/ 22.78 GFLOPS | Progress: (16/20) | 12.18 s
[Task 22/25] Current/Best: 10.55/ 22.78 GFLOPS | Progress: (20/20) | 15.28 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 2.39/ 20.55 GFLOPS | Progress: (4/20) | 6.99 s
[Task 23/25] Current/Best: 1.55/ 23.89 GFLOPS | Progress: (8/20) | 11.92 s
[Task 23/25] Current/Best: 10.25/ 23.89 GFLOPS | Progress: (12/20) | 17.32 s
[Task 23/25] Current/Best: 22.59/ 23.89 GFLOPS | Progress: (16/20) | 22.92 s
[Task 23/25] Current/Best: 3.08/ 23.89 GFLOPS | Progress: (20/20) | 26.33 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 1.55/ 6.03 GFLOPS | Progress: (4/20) | 13.87 s
[Task 24/25] Current/Best: 3.61/ 6.03 GFLOPS | Progress: (8/20) | 16.97 s
[Task 24/25] Current/Best: 3.82/ 7.11 GFLOPS | Progress: (12/20) | 18.60 s
[Task 24/25] Current/Best: 6.77/ 10.01 GFLOPS | Progress: (16/20) | 25.62 s
[Task 24/25] Current/Best: 3.64/ 10.01 GFLOPS | Progress: (20/20) | 36.61 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+ Done.
+
[Task 25/25] Current/Best: 3.00/ 9.72 GFLOPS | Progress: (4/20) | 8.53 s
[Task 25/25] Current/Best: 8.85/ 9.72 GFLOPS | Progress: (8/20) | 19.52 s
[Task 25/25] Current/Best: 7.36/ 9.72 GFLOPS | Progress: (12/20) | 22.86 s
[Task 25/25] Current/Best: 2.97/ 9.72 GFLOPS | Progress: (16/20) | 28.51 s
[Task 25/25] Current/Best: 5.56/ 9.72 GFLOPS | Progress: (20/20) | 35.66 s
@@ -666,8 +663,8 @@ Verify that the optimized model runs and produces the same results:
.. code-block:: none
- class='n02123045 tabby, tabby cat' with probability=0.621104
- class='n02123159 tiger cat' with probability=0.356378
+ class='n02123045 tabby, tabby cat' with probability=0.621103
+ class='n02123159 tiger cat' with probability=0.356379
class='n02124075 Egyptian cat' with probability=0.019712
class='n02129604 tiger, Panthera tigris' with probability=0.001215
class='n04040759 radiator' with probability=0.000262
@@ -724,8 +721,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 411.87878949999686, 'median': 409.846005299994, 'std': 4.990028647976033}
- unoptimized: {'mean': 513.497117569998, 'median': 513.9784247000023, 'std': 2.379501423445631}
+ optimized: {'mean': 412.0973418800054, 'median': 412.5166654000054, 'std': 1.5141097041492024}
+ unoptimized: {'mean': 512.8374819899979, 'median': 513.379893900003, 'std': 2.048731521479577}
@@ -748,7 +745,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 12 minutes 52.874 seconds)
+ **Total running time of the script:** ( 12 minutes 21.049 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 cfb8b97601..6c9783930a 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -274,7 +274,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.324e-07 secs/op
+ 1.377e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index d37c82262e..9d1afe3a67 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -270,7 +270,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0xc2e33f0)), stage(b, placeholder(b, 0x8ea8c50)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.R [...]
+ [stage(a, placeholder(a, 0xdfead00)), stage(b, placeholder(b, 0x10038f90)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T. [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index 217082940c..f774a768f1 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,32 +5,32 @@
Computation times
=================
-**16:32.281** total execution time for **tutorial** files:
+**15:56.194** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 12:52.874 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 12:21.049 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:32.083 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:24.282 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:02.713 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:02.551 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:37.045 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:37.875 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:25.182 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:27.550 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.349 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.825 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.851 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.864 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.185 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.198 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.000 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.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_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 68e5c97001..2c7b0d983c 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -285,7 +285,7 @@ helper function to run a profile of the TVM generated code.
.. code-block:: none
- Numpy running time: 0.000007
+ Numpy running time: 0.000008
naive: 0.000007
@@ -498,10 +498,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 6.619640000735671e-06 1.0
- naive 6.9131e-06 1.0443317158080674
- parallel 7.0763e-06 1.068985624477098
- vector 2.45975e-05 3.7158365103338498
+ numpy 7.911820000572334e-06 1.0
+ naive 6.7071000000000005e-06 0.8477316217399807
+ parallel 7.0035e-06 0.8851945569405488
+ vector 2.48371e-05 3.1392397701418013
@@ -922,7 +922,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.018353
+ Numpy running time: 0.019915
@@ -980,7 +980,7 @@ optimizations.
.. code-block:: none
- none: 3.558139
+ none: 3.473186
@@ -1080,7 +1080,7 @@ schedule.
.. code-block:: none
- blocking: 0.295713
+ blocking: 0.323740
@@ -1164,7 +1164,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.335590
+ vectorization: 0.353062
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1230,7 +1230,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.118354
+ loop permutation: 0.121582
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1321,7 +1321,7 @@ optimized schedule.
.. code-block:: none
- array packing: 0.109168
+ array packing: 0.108879
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1404,7 +1404,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.111663
+ block caching: 0.111291
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1478,7 +1478,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.147037
+ parallelization: 0.147603
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1548,13 +1548,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.5581392619999996 1.0
- blocking 0.2957133518 0.08310898759869845
- vectorization 0.33559017090000004 0.09431619905494303
- loop permutation 0.118353562 0.03326276834186486
- array packing 0.10916753819999998 0.03068107518046886
- block caching 0.11166263349999998 0.03138231116823555
- parallelization 0.14703718289999998 0.04132417875554192
+ none 3.4731861198 1.0
+ blocking 0.3237404139 0.09321136349544157
+ vectorization 0.3530619727 0.10165362883585713
+ loop permutation 0.12158248149999999 0.035006036908555076
+ array packing 0.10887909850000002 0.03134847795207408
+ block caching 0.11129050930000002 0.032042771524840884
+ parallelization 0.1476034903 0.04249800765312848
@@ -1596,7 +1596,7 @@ the computation for specific platforms.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 2.713 seconds)
+ **Total running time of the script:** ( 1 minutes 2.551 seconds)
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index 4fd00b9f34..a995f521ab 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-3f56a95b872ca1078b209db9d8d368c8a29c0923
+e5ae4347ddf492a9e017dd1dba47df364a27a197
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index b4f85197cd..f4cc9a8431 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -590,7 +590,7 @@ class:['truck 0.9266'] left:471 top:83 right:689 bottom:169
class:['bicycle 0.9984'] left:111 top:113 right:577 bottom:447
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 19.896 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 23.554 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-darknet-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7716f96385bd5abb6e822041e285be54/from_darknet.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_darknet.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 9fa57eb20b..024bfb88a5 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -444,7 +444,7 @@
<span class="nb">print</span><span class="p">(</span><span class="s2">"x"</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">x</span><span class="o">.</span><span class="n">shape</span></a><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip0132a9fd-574d-4c94-a24a-11934417a0f0 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.zip0345609a-aa53-4983-888b-c0a10eb99acb 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 98c9327a1a..a7a07147df 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -454,13 +454,13 @@ Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdo
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
0%| | 0.00/41.5M [00:00<?, ?B/s]
- 19%|#9 | 7.99M/41.5M [00:00<00:00, 58.0MB/s]
- 35%|###4 | 14.3M/41.5M [00:00<00:00, 47.5MB/s]
- 54%|#####3 | 22.3M/41.5M [00:00<00:00, 58.0MB/s]
- 68%|######7 | 28.1M/41.5M [00:00<00:00, 53.5MB/s]
- 80%|######## | 33.3M/41.5M [00:00<00:00, 50.0MB/s]
- 96%|#########6| 40.0M/41.5M [00:00<00:00, 51.5MB/s]
-100%|##########| 41.5M/41.5M [00:00<00:00, 53.6MB/s]
+ 15%|#5 | 6.33M/41.5M [00:00<00:00, 66.1MB/s]
+ 30%|### | 12.6M/41.5M [00:00<00:00, 51.6MB/s]
+ 43%|####2 | 17.8M/41.5M [00:00<00:00, 33.7MB/s]
+ 58%|#####7 | 24.0M/41.5M [00:00<00:00, 35.9MB/s]
+ 77%|#######7 | 32.0M/41.5M [00:00<00:00, 46.0MB/s]
+ 92%|#########2| 38.4M/41.5M [00:00<00:00, 51.2MB/s]
+100%|##########| 41.5M/41.5M [00:00<00:00, 46.5MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index ed4e175162..3968a87a39 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -437,10 +437,12 @@ be unstable.</p>
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
0%| | 0.00/44.7M [00:00<?, ?B/s]
- 27%|##7 | 12.1M/44.7M [00:00<00:00, 127MB/s]
- 54%|#####4 | 24.2M/44.7M [00:00<00:00, 107MB/s]
- 77%|#######7 | 34.6M/44.7M [00:00<00:00, 106MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 106MB/s]
+ 18%|#7 | 7.99M/44.7M [00:00<00:00, 76.6MB/s]
+ 40%|#### | 18.1M/44.7M [00:00<00:00, 86.7MB/s]
+ 59%|#####8 | 26.3M/44.7M [00:00<00:00, 79.6MB/s]
+ 76%|#######5 | 33.9M/44.7M [00:00<00:00, 78.7MB/s]
+ 96%|#########5| 42.8M/44.7M [00:00<00:00, 83.6MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 74.4MB/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 30df0c9964..85c03f65bf 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -657,7 +657,7 @@ banana (score = 0.00022)
desk (score = 0.00019)
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 31.401 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 35.258 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 a9d3d6eea7..cc532e22e2 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>06:49.101</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>07:04.515</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -354,43 +354,43 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></td>
-<td><p>01:31.401</p></td>
+<td><p>01:35.258</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></td>
-<td><p>01:19.896</p></td>
+<td><p>01:23.554</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_paddle.html#sphx-glr-how-to-compile-models-from-paddle-py"><span class="std std-ref">Compile PaddlePaddle Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_paddle.py</span></code>)</p></td>
-<td><p>00:57.247</p></td>
+<td><p>00:59.622</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>00:38.090</p></td>
+<td><p>00:39.309</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>00:32.153</p></td>
+<td><p>00:33.471</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>00:30.114</p></td>
+<td><p>00:31.014</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>00:28.133</p></td>
+<td><p>00:28.503</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>00:26.369</p></td>
+<td><p>00:27.353</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>00:22.958</p></td>
+<td><p>00:23.635</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>00:02.738</p></td>
+<td><p>00:02.796</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/deploy_models/deploy_model_on_adreno.html b/docs/how_to/deploy_models/deploy_model_on_adreno.html
index 6b687f0250..4676140e85 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno.html
@@ -799,7 +799,7 @@ Top5 predictions:
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 2542.2163 2541.2295 2547.7983 2540.4597 2.4130
+ 2750.3544 2750.1791 2754.7262 2747.9918 1.7600
</pre></div>
</div>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-model-on-adreno-py">
diff --git a/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html b/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
index 0388239803..121556d85c 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
@@ -443,26 +443,30 @@ to run this tutorial with a real device over rpc.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5
8192/102967424 [..............................] - ETA: 0s
- 7725056/102967424 [=>............................] - ETA: 0s
- 12730368/102967424 [==>...........................] - ETA: 0s
- 17481728/102967424 [====>.........................] - ETA: 0s
- 22609920/102967424 [=====>........................] - ETA: 0s
- 26034176/102967424 [======>.......................] - ETA: 0s
- 32751616/102967424 [========>.....................] - ETA: 0s
- 35971072/102967424 [=========>....................] - ETA: 0s
- 42573824/102967424 [===========>..................] - ETA: 0s
- 47718400/102967424 [============>.................] - ETA: 0s
+ 4775936/102967424 [>.............................] - ETA: 1s
+ 8380416/102967424 [=>............................] - ETA: 1s
+ 10575872/102967424 [==>...........................] - ETA: 1s
+ 15024128/102967424 [===>..........................] - ETA: 1s
+ 16769024/102967424 [===>..........................] - ETA: 1s
+ 25157632/102967424 [======>.......................] - ETA: 1s
+ 33546240/102967424 [========>.....................] - ETA: 1s
+ 40189952/102967424 [==========>...................] - ETA: 0s
+ 41934848/102967424 [===========>..................] - ETA: 1s
+ 47046656/102967424 [============>.................] - ETA: 0s
50323456/102967424 [=============>................] - ETA: 0s
- 58023936/102967424 [===============>..............] - ETA: 0s
- 63528960/102967424 [=================>............] - ETA: 0s
- 68501504/102967424 [==================>...........] - ETA: 0s
- 73588736/102967424 [====================>.........] - ETA: 0s
- 78610432/102967424 [=====================>........] - ETA: 0s
- 83623936/102967424 [=======================>......] - ETA: 0s
- 88727552/102967424 [========================>.....] - ETA: 0s
- 93495296/102967424 [==========================>...] - ETA: 0s
- 98705408/102967424 [===========================>..] - ETA: 0s
-102967424/102967424 [==============================] - 1s 0us/step
+ 56967168/102967424 [===============>..............] - ETA: 0s
+ 58712064/102967424 [================>.............] - ETA: 0s
+ 65323008/102967424 [==================>...........] - ETA: 0s
+ 67305472/102967424 [==================>...........] - ETA: 0s
+ 69296128/102967424 [===================>..........] - ETA: 0s
+ 75489280/102967424 [====================>.........] - ETA: 0s
+ 82878464/102967424 [=======================>......] - ETA: 0s
+ 83877888/102967424 [=======================>......] - ETA: 0s
+ 90521600/102967424 [=========================>....] - ETA: 0s
+ 92266496/102967424 [=========================>....] - ETA: 0s
+ 98910208/102967424 [===========================>..] - ETA: 0s
+102850560/102967424 [============================>.] - ETA: 0s
+102967424/102967424 [==============================] - 2s 0us/step
</pre></div>
</div>
</div>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 91c4cf27a9..7cbd333a1e 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -667,7 +667,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.6415 15.6379 15.7834 15.5298 0.0603
+ 16.1671 16.1450 16.5010 15.9790 0.1270
</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 be63ac5c1b..92f6630a77 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -459,21 +459,31 @@ be unstable.</p>
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
0%| | 0.00/170M [00:00<?, ?B/s]
- 5%|4 | 8.12M/170M [00:00<00:03, 51.5MB/s]
- 13%|#3 | 22.6M/170M [00:00<00:01, 81.8MB/s]
- 18%|#7 | 30.5M/170M [00:00<00:01, 77.6MB/s]
- 25%|##4 | 42.1M/170M [00:00<00:01, 91.4MB/s]
- 34%|###3 | 57.7M/170M [00:00<00:01, 114MB/s]
- 42%|####2 | 72.0M/170M [00:00<00:00, 122MB/s]
- 49%|####9 | 84.0M/170M [00:00<00:00, 105MB/s]
- 57%|#####6 | 96.0M/170M [00:00<00:00, 109MB/s]
- 64%|######4 | 109M/170M [00:01<00:00, 118MB/s]
- 71%|#######1 | 121M/170M [00:01<00:00, 111MB/s]
- 78%|#######7 | 132M/170M [00:01<00:00, 105MB/s]
- 84%|########4 | 143M/170M [00:01<00:00, 107MB/s]
- 90%|######### | 153M/170M [00:01<00:00, 105MB/s]
- 96%|#########6| 164M/170M [00:01<00:00, 90.5MB/s]
-100%|##########| 170M/170M [00:01<00:00, 101MB/s]
+ 4%|3 | 6.30M/170M [00:00<00:02, 61.9MB/s]
+ 7%|7 | 12.2M/170M [00:00<00:02, 57.3MB/s]
+ 12%|#1 | 19.8M/170M [00:00<00:02, 66.9MB/s]
+ 15%|#5 | 26.2M/170M [00:00<00:02, 53.9MB/s]
+ 19%|#8 | 32.0M/170M [00:00<00:02, 50.7MB/s]
+ 24%|##3 | 40.0M/170M [00:00<00:02, 52.2MB/s]
+ 28%|##8 | 48.0M/170M [00:00<00:02, 55.5MB/s]
+ 33%|###2 | 56.0M/170M [00:01<00:02, 57.5MB/s]
+ 37%|###6 | 62.3M/170M [00:01<00:01, 58.4MB/s]
+ 40%|#### | 68.1M/170M [00:01<00:01, 59.0MB/s]
+ 43%|####3 | 73.8M/170M [00:01<00:01, 54.7MB/s]
+ 47%|####7 | 80.0M/170M [00:01<00:01, 53.6MB/s]
+ 51%|##### | 86.3M/170M [00:01<00:01, 55.2MB/s]
+ 54%|#####3 | 91.6M/170M [00:01<00:01, 47.5MB/s]
+ 57%|#####6 | 96.3M/170M [00:02<00:02, 34.5MB/s]
+ 61%|######1 | 104M/170M [00:02<00:01, 40.7MB/s]
+ 66%|######5 | 112M/170M [00:02<00:01, 43.0MB/s]
+ 71%|####### | 120M/170M [00:02<00:01, 46.7MB/s]
+ 75%|#######5 | 128M/170M [00:02<00:00, 50.3MB/s]
+ 80%|######## | 136M/170M [00:02<00:00, 54.8MB/s]
+ 85%|########4 | 144M/170M [00:02<00:00, 56.6MB/s]
+ 89%|########9 | 152M/170M [00:03<00:00, 54.0MB/s]
+ 93%|#########3| 158M/170M [00:03<00:00, 55.7MB/s]
+ 97%|#########6| 165M/170M [00:03<00:00, 58.2MB/s]
+100%|##########| 170M/170M [00:03<00:00, 51.9MB/s]
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode=& [...]
@@ -571,7 +581,7 @@ 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> ( 3 minutes 37.138 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 46.558 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 2e800847aa..9e78af6c0e 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -500,8 +500,8 @@ training. Other models require a full post training calibration.</p>
Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
0%| | 0.00/13.6M [00:00<?, ?B/s]
- 92%|#########2| 12.5M/13.6M [00:00<00:00, 131MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 135MB/s]
+ 59%|#####8 | 7.99M/13.6M [00:00<00:00, 38.6MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 58.9MB/s]
</pre></div>
</div>
</div>
@@ -592,7 +592,7 @@ output values are identical out of 1000 outputs from mobilenet v2.</p>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.7204 90.6843 92.1979 90.2426 0.3522
+ 88.9308 88.2800 91.3655 88.0628 0.9692
</pre></div>
</div>
<div class="admonition note">
@@ -631,7 +631,7 @@ This includes support for the VNNI 8 bit dot product instruction (CascadeLake or
<div class="section" id="deploy-a-quantized-tflite-model">
<h2>Deploy a quantized TFLite Model<a class="headerlink" href="#deploy-a-quantized-tflite-model" title="Permalink to this headline">¶</a></h2>
<p>TODO</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 17.958 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 19.728 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 29e54d2223..0c6cf17cf8 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -585,7 +585,7 @@ TFLite Top-5 labels: [387 102 386 341 349]
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 121.0102 120.9625 123.1915 119.6895 0.5799
+ 121.0844 121.2201 122.4309 118.6074 0.7515
</pre></div>
</div>
<div class="admonition note">
@@ -613,7 +613,7 @@ network for ARM CPU</span></a>.</p></li>
</ul>
</div></blockquote>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 35.394 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 29.938 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 4f8332a3b2..1ecfb7df10 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -526,7 +526,7 @@ for calibration. But the accuracy might be impacted.</p>
DeprecationWarning,
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 35.606 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 39.936 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 3fbdf94b7d..7cef05f46f 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -468,24 +468,25 @@ 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]
- 1%| | 1194/132723 [00:00<00:11, 11915.01KB/s]
- 3%|2 | 3786/132723 [00:00<00:06, 20058.73KB/s]
- 7%|7 | 9509/132723 [00:00<00:03, 36981.11KB/s]
- 13%|#3 | 17878/132723 [00:00<00:02, 55386.06KB/s]
- 20%|#9 | 26291/132723 [00:00<00:01, 65737.89KB/s]
- 26%|##6 | 34665/132723 [00:00<00:01, 71850.50KB/s]
- 33%|###2 | 43164/132723 [00:00<00:01, 76142.21KB/s]
- 39%|###8 | 51232/132723 [00:00<00:01, 77583.45KB/s]
- 45%|####5 | 59733/132723 [00:00<00:00, 79902.00KB/s]
- 51%|#####1 | 68271/132723 [00:01<00:00, 81589.82KB/s]
- 58%|#####7 | 76639/132723 [00:01<00:00, 82227.57KB/s]
- 64%|######4 | 85076/132723 [00:01<00:00, 82877.69KB/s]
- 70%|####### | 93520/132723 [00:01<00:00, 83347.81KB/s]
- 77%|#######6 | 101994/132723 [00:01<00:00, 83766.91KB/s]
- 83%|########3 | 110500/132723 [00:01<00:00, 84154.44KB/s]
- 90%|########9 | 118916/132723 [00:01<00:00, 84030.66KB/s]
- 96%|#########5| 127413/132723 [00:01<00:00, 84309.45KB/s]
-100%|##########| 132723/132723 [00:01<00:00, 75173.46KB/s]
+ 1%| | 1308/132723 [00:00<00:10, 13040.62KB/s]
+ 3%|3 | 4237/132723 [00:00<00:05, 22586.65KB/s]
+ 8%|7 | 10121/132723 [00:00<00:03, 39126.63KB/s]
+ 12%|#2 | 16441/132723 [00:00<00:02, 48624.36KB/s]
+ 18%|#7 | 23636/132723 [00:00<00:01, 57032.13KB/s]
+ 23%|##3 | 31133/132723 [00:00<00:01, 63125.25KB/s]
+ 29%|##9 | 39038/132723 [00:00<00:01, 68327.34KB/s]
+ 35%|###5 | 46987/132723 [00:00<00:01, 71876.62KB/s]
+ 41%|####1 | 54929/132723 [00:00<00:01, 74232.14KB/s]
+ 47%|####7 | 62701/132723 [00:01<00:00, 75299.27KB/s]
+ 53%|#####3 | 70589/132723 [00:01<00:00, 76390.87KB/s]
+ 59%|#####9 | 78560/132723 [00:01<00:00, 77398.99KB/s]
+ 65%|######5 | 86458/132723 [00:01<00:00, 77874.69KB/s]
+ 71%|#######1 | 94448/132723 [00:01<00:00, 78484.07KB/s]
+ 77%|#######7 | 102380/132723 [00:01<00:00, 78735.08KB/s]
+ 83%|########3 | 110273/132723 [00:01<00:00, 78790.94KB/s]
+ 89%|########9 | 118255/132723 [00:01<00:00, 79098.68KB/s]
+ 95%|#########5| 126172/132723 [00:01<00:00, 79118.26KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 70493.20KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -524,7 +525,7 @@ 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> ( 3 minutes 51.147 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> ( 3 minutes 59.099 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 3f1c105d4a..356aa10380 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>16:23.800</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>16:49.430</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -354,47 +354,47 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></td>
-<td><p>03:51.147</p></td>
+<td><p>03:59.099</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></td>
-<td><p>03:37.138</p></td>
+<td><p>03:46.558</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></td>
-<td><p>02:35.394</p></td>
+<td><p>02:29.938</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_quantized.html#sphx-glr-how-to-deploy-models-deploy-quantized-py"><span class="std std-ref">Deploy a Quantized Model on Cuda</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_quantized.py</span></code>)</p></td>
-<td><p>01:35.606</p></td>
+<td><p>01:39.936</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>01:17.958</p></td>
+<td><p>01:19.728</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_adreno.html#sphx-glr-how-to-deploy-models-deploy-model-on-adreno-py"><span class="std std-ref">Deploy the Pretrained Model on Adreno™</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_adreno.py</span></code>)</p></td>
-<td><p>00:54.873</p></td>
+<td><p>00:58.394</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_adreno_tvmc.html#sphx-glr-how-to-deploy-models-deploy-model-on-adreno-tvmc-py"><span class="std std-ref">Deploy the Pretrained Model on Adreno™ with tvmc Interface</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_adreno_tvmc.py</span></code>)</p></td>
-<td><p>00:51.262</p></td>
+<td><p>00:53.202</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>00:42.726</p></td>
+<td><p>00:44.689</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_nano.html#sphx-glr-how-to-deploy-models-deploy-model-on-nano-py"><span class="std std-ref">Deploy the Pretrained Model on Jetson Nano</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_nano.py</span></code>)</p></td>
-<td><p>00:28.962</p></td>
+<td><p>00:29.210</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></td>
-<td><p>00:28.724</p></td>
+<td><p>00:28.670</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></td>
-<td><p>00:00.010</p></td>
+<td><p>00:00.006</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/extend_tvm/bring_your_own_datatypes.html b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
index ca68b8a961..f4c346d721 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -624,7 +624,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
<span class="n">module</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">params</span></a> <span class="o">=</span> <span class="n">get_mobilenet</span><span class="p">()</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipb5ee551d-6d6e-49f4-9bed-cc3ed7b09e10 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.zip929e5e9d-8d88-4ea5-95fc-5c8e6ad4ee23 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
</pre></div>
</div>
<p>It’s easy to execute MobileNet with native TVM:</p>
diff --git a/docs/how_to/extend_tvm/sg_execution_times.html b/docs/how_to/extend_tvm/sg_execution_times.html
index fc7c72bf09..4d4d60160a 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-extend-tvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:54.081</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:58.769</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,19 +354,19 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></td>
-<td><p>00:50.293</p></td>
+<td><p>00:54.661</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>00:02.725</p></td>
+<td><p>00:02.952</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></td>
-<td><p>00:01.056</p></td>
+<td><p>00:01.149</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></td>
-<td><p>00:00.007</p></td>
+<td><p>00:00.008</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index a2af218aa7..3c06a1b48a 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -531,10 +531,10 @@ profile the execution time of each passes.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 22339us [22339us] (48.07%; 48.07%)
-FoldScaleAxis: 24132us [7us] (51.93%; 51.93%)
- FoldConstant: 24125us [1720us] (51.91%; 99.97%)
- InferType: 22405us [22405us] (48.21%; 92.87%)
+InferType: 23565us [23565us] (49.07%; 49.07%)
+FoldScaleAxis: 24454us [9us] (50.93%; 50.93%)
+ FoldConstant: 24444us [1762us] (50.91%; 99.96%)
+ InferType: 22683us [22683us] (47.24%; 92.79%)
</pre></div>
</div>
</div>
@@ -556,10 +556,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 21709us [21709us] (48.40%; 48.40%)
-FoldScaleAxis: 23141us [5us] (51.60%; 51.60%)
- FoldConstant: 23136us [1729us] (51.59%; 99.98%)
- InferType: 21407us [21407us] (47.73%; 92.53%)
+InferType: 23804us [23804us] (49.22%; 49.22%)
+FoldScaleAxis: 24559us [8us] (50.78%; 50.78%)
+ FoldConstant: 24551us [1875us] (50.76%; 99.97%)
+ InferType: 22675us [22675us] (46.89%; 92.36%)
</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 201b377caa..59be08adbd 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -580,7 +580,7 @@ latency of convolution.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Convolution: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">*</span> <span cl [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 53.507263 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 53.512542 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 16a6a19dd6..1b2c741221 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -862,7 +862,7 @@ be able to run on our build server</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"conv2d with tensor core: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">* [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 11.759011 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 12.223469 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 5ba4c9050d..ee74ff08b2 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -477,8 +477,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
<span class="nb">print</span><span class="p">(</span><span class="s2">"Baseline: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018365
-Baseline: 3.399232
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019328
+Baseline: 3.498159
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -537,7 +537,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt1: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.299479
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.329814
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -594,7 +594,7 @@ vastly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt2: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.334021
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.353135
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -649,7 +649,7 @@ the access pattern for A matrix is more cache friendly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt3: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.117414
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.134273
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -726,7 +726,7 @@ flattening.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt4: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.109954
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.110898
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -804,7 +804,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt5: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111315
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.112132
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -884,7 +884,7 @@ class Module:
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt6: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">opt6_time</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.147122
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.148916
</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 7eb2221628..ed7dbe98f3 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:35.456</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:36.796</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,15 +354,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></td>
-<td><p>00:32.400</p></td>
+<td><p>00:33.689</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>00:01.930</p></td>
+<td><p>00:01.894</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>00:01.127</p></td>
+<td><p>00:01.212</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 22042f2176..73626f2c03 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autoscheduler-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>10:20.424</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>10:36.831</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -354,27 +354,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></td>
-<td><p>06:25.850</p></td>
+<td><p>06:28.831</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>01:42.813</p></td>
+<td><p>01:47.644</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_network_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-cuda-py"><span class="std std-ref">Auto-scheduling a Neural Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_cuda.py</span></code>)</p></td>
-<td><p>01:10.653</p></td>
+<td><p>01:13.908</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>00:33.748</p></td>
+<td><p>00:36.900</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></td>
-<td><p>00:13.971</p></td>
+<td><p>00:15.119</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></td>
-<td><p>00:13.389</p></td>
+<td><p>00:14.429</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 4f7cbcae0c..21e9da13dc 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
@@ -494,9 +494,6 @@ file and apply it.</p>
<span class="k">del</span> <span class="n">measure_ctx</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>.T
-</pre></div>
-</div>
<p>We can lower the schedule to see the IR after auto-scheduling.
The auto-scheduler correctly performs optimizations including multi-level tiling,
cooperative fetching, unrolling and operator fusion.</p>
@@ -513,664 +510,86 @@ class Module:
@T.prim_func
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True})
- blockIdx_x = T.launch_thread("blockIdx.x", 16)
- conv2d_nchw = T.allocate([28], "float32", "local")
- pad_temp_shared = T.allocate([1296], "float32", "shared")
- kernel_shared = T.allocate([4608], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 56)
- conv2d_nchw_1 = T.Buffer((16,), data=conv2d_nchw, scope="local", align=16)
+ blockIdx_x = T.launch_thread("blockIdx.x", 32)
+ conv2d_nchw = T.allocate([7], "float32", "local")
+ pad_temp_shared = T.allocate([1568], "float32", "shared")
+ kernel_shared = T.allocate([512], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 112)
+ conv2d_nchw_1 = T.Buffer((7,), data=conv2d_nchw, scope="local", align=16)
conv2d_nchw_1[0] = T.float32(0)
- conv2d_nchw_1[4] = T.float32(0)
- conv2d_nchw_1[8] = T.float32(0)
- conv2d_nchw_1[12] = T.float32(0)
- conv2d_nchw_1[16] = T.float32(0)
- conv2d_nchw_1[20] = T.float32(0)
- conv2d_nchw_1[24] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
- conv2d_nchw_1[5] = T.float32(0)
- conv2d_nchw_1[9] = T.float32(0)
- conv2d_nchw_1[13] = T.float32(0)
- conv2d_nchw_1[17] = T.float32(0)
- conv2d_nchw_1[21] = T.float32(0)
- conv2d_nchw_1[25] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
- conv2d_nchw_1[6] = T.float32(0)
- conv2d_nchw_1[10] = T.float32(0)
- conv2d_nchw_1[14] = T.float32(0)
- conv2d_nchw_1[18] = T.float32(0)
- conv2d_nchw_1[22] = T.float32(0)
- conv2d_nchw_1[26] = T.float32(0)
conv2d_nchw_1[3] = T.float32(0)
- conv2d_nchw_1[7] = T.float32(0)
- conv2d_nchw_1[11] = T.float32(0)
- conv2d_nchw_1[15] = T.float32(0)
- conv2d_nchw_1[19] = T.float32(0)
- conv2d_nchw_1[23] = T.float32(0)
- conv2d_nchw_1[27] = T.float32(0)
- for rc_outer_outer in range(32):
- cse_var_2: T.int32 = rc_outer_outer * 784
- cse_var_1: T.int32 = rc_outer_outer * 144
+ conv2d_nchw_1[4] = T.float32(0)
+ conv2d_nchw_1[5] = T.float32(0)
+ conv2d_nchw_1[6] = T.float32(0)
+ for rc_outer_outer, ry_outer_outer, rx_outer_outer in T.grid(16, 3, 3):
+ cse_var_2: T.int32 = rc_outer_outer * 288
+ cse_var_1: T.int32 = ry_outer_outer * 3
threadIdx_x_1 = T.env_thread("threadIdx.x")
- pad_temp_shared_1 = T.Buffer((1296,), data=pad_temp_shared, scope="shared")
+ pad_temp_shared_1 = T.Buffer((1568,), data=pad_temp_shared, scope="shared")
data_1 = T.Buffer((25088,), data=data.data)
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(9 <= threadIdx_x_1 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + threadIdx_x_1 // 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 56] = T.if_then_else(9 <= (threadIdx_x_1 + 56) % 81 and (threadIdx_x_1 + 56) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 56) // 81 * 49 + (threadIdx_x_1 + 56) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 112] = T.if_then_else(9 <= (threadIdx_x_1 + 31) % 81 and (threadIdx_x_1 + 31) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 112) // 81 * 49 + (threadIdx_x_1 + 31) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 168] = T.if_then_else(3 <= threadIdx_x_1 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 168) // 81 * 49 + (threadIdx_x_1 + 6) // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 224] = T.if_then_else(9 <= (threadIdx_x_1 + 62) % 81 and (threadIdx_x_1 + 62) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 224) // 81 * 49 + (threadIdx_x_1 + 62) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 280] = T.if_then_else(9 <= (threadIdx_x_1 + 37) % 81 and (threadIdx_x_1 + 37) % 81 < 72 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 280) // 81 * 49 + (threadIdx_x_1 + 37) % 81 // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 336] = T.if_then_else(1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 336) // 81 * 49 + (threadIdx_x_1 + 12) // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 392] = T.if_then_else(9 <= (threadIdx_x_1 + 68) % 81 and (threadIdx_x_1 + 68) % 81 < 72 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 392) // 81 * 49 + (threadIdx_x_1 + 68) % 81 // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 448] = T.if_then_else(9 <= (threadIdx_x_1 + 43) % 81 and (threadIdx_x_1 + 43) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 448) // 81 * 49 + (threadIdx_x_1 + 43) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 504] = T.if_then_else(threadIdx_x_1 < 54 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 504) // 81 * 49 + threadIdx_x_1 // 9 * 7 + threadIdx_x_1 % 9 + 6], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 560] = T.if_then_else(9 <= (threadIdx_x_1 + 74) % 81 and (threadIdx_x_1 + 74) % 81 < 72 and 1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 560) // 81 * 49 + (threadIdx_x_1 + 74) % 81 // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 616] = T.if_then_else(9 <= (threadIdx_x_1 + 49) % 81 and (threadIdx_x_1 + 49) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 616) // 81 * 49 + (threadIdx_x_1 + 49) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 672] = T.if_then_else(threadIdx_x_1 < 48 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 672) // 81 * 49 + (threadIdx_x_1 + 24) // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 728] = T.if_then_else(9 <= (threadIdx_x_1 + 80) % 81 and (threadIdx_x_1 + 80) % 81 < 72 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 728) // 81 * 49 + (threadIdx_x_1 + 80) % 81 // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 784] = T.if_then_else(9 <= (threadIdx_x_1 + 55) % 81 and (threadIdx_x_1 + 55) % 81 < 72 and 1 <= (threadIdx_x_1 + 1) % 9 and (threadIdx_x_1 + 1) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 784) // 81 * 49 + (threadIdx_x_1 + 55) % 81 // 9 * 7 + (threadIdx_x_1 + 1) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 840] = T.if_then_else(9 <= (threadIdx_x_1 + 30) % 81 and (threadIdx_x_1 + 30) % 81 < 72 and 1 <= (threadIdx_x_1 + 3) % 9 and (threadIdx_x_1 + 3) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 840) // 81 * 49 + (threadIdx_x_1 + 30) % 81 // 9 * 7 + (threadIdx_x_1 + 3) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 896] = T.if_then_else(4 <= threadIdx_x_1 and 1 <= (threadIdx_x_1 + 5) % 9 and (threadIdx_x_1 + 5) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 896) // 81 * 49 + (threadIdx_x_1 + 5) // 9 * 7 + (threadIdx_x_1 + 5) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 952] = T.if_then_else(9 <= (threadIdx_x_1 + 61) % 81 and (threadIdx_x_1 + 61) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 952) // 81 * 49 + (threadIdx_x_1 + 61) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1008] = T.if_then_else(1 <= (threadIdx_x_1 // 9 + 4) % 9 and (threadIdx_x_1 + 36) % 81 < 72 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1008) // 81 * 49 + (threadIdx_x_1 // 9 + 4) % 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1064] = T.if_then_else(1 <= (threadIdx_x_1 + 2) % 9 and (threadIdx_x_1 + 2) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1064) // 81 * 49 + (threadIdx_x_1 + 11) // 9 * 7 + (threadIdx_x_1 + 2) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1120] = T.if_then_else(9 <= (threadIdx_x_1 + 67) % 81 and (threadIdx_x_1 + 67) % 81 < 72 and 1 <= (threadIdx_x_1 + 4) % 9 and (threadIdx_x_1 + 4) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1120) // 81 * 49 + (threadIdx_x_1 + 67) % 81 // 9 * 7 + (threadIdx_x_1 + 4) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1176] = T.if_then_else(9 <= (threadIdx_x_1 + 42) % 81 and (threadIdx_x_1 + 42) % 81 < 72 and 1 <= (threadIdx_x_1 + 6) % 9 and (threadIdx_x_1 + 6) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1176) // 81 * 49 + (threadIdx_x_1 + 42) % 81 // 9 * 7 + (threadIdx_x_1 + 6) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- pad_temp_shared_1[threadIdx_x_1 + 1232] = T.if_then_else(threadIdx_x_1 < 55 and 1 <= (threadIdx_x_1 + 8) % 9 and (threadIdx_x_1 + 8) % 9 < 8, data_1[cse_var_2 + (threadIdx_x_1 + 1232) // 81 * 49 + (threadIdx_x_1 + 17) // 9 * 7 + (threadIdx_x_1 + 8) % 9 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 56):
- if T.likely(threadIdx_x_1 < 8):
- pad_temp_shared_1[threadIdx_x_1 + 1288] = T.float32(0)
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(1 <= threadIdx_x_1 % 49 // 7 + ry_outer_outer and threadIdx_x_1 % 49 // 7 + ry_outer_outer < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer - 8], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 112] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 2) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 2) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 104], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 224] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 4) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 4) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 216], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 336] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 6) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 6) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 328], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 448] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 1) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 1) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 440], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 560] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 3) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 3) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 552], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 672] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 5) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 5) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 664], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 784] = T.if_then_else(1 <= threadIdx_x_1 % 49 // 7 + ry_outer_outer and threadIdx_x_1 % 49 // 7 + ry_outer_outer < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 776], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 896] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 2) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 2) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 888], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1008] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 4) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 4) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1000], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1120] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 6) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 6) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1112], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1232] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 1) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 1) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1224], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1344] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 3) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 3) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1336], T.float32(0))
+ with T.launch_thread(threadIdx_x_1, 112):
+ pad_temp_shared_1[threadIdx_x_1 + 1456] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + 5) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + 5) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 1568 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer + 1448], T.float32(0))
threadIdx_x_2 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((4608,), data=kernel_shared, scope="shared")
+ kernel_shared_1 = T.Buffer((512,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8]
- kernel_shared_1[threadIdx_x_2 * 8 + 1] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 1]
- kernel_shared_1[threadIdx_x_2 * 8 + 2] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 2]
- kernel_shared_1[threadIdx_x_2 * 8 + 3] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 3]
- kernel_shared_1[threadIdx_x_2 * 8 + 4] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 4]
- kernel_shared_1[threadIdx_x_2 * 8 + 5] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 5]
- kernel_shared_1[threadIdx_x_2 * 8 + 6] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 6]
- kernel_shared_1[threadIdx_x_2 * 8 + 7] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 7]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 448] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 16) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 449] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 17) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 450] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 18) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 451] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 19) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 452] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 20) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 453] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 21) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 454] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 22) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 455] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 56) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 23) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 896] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 32) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 897] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 33) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 898] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 34) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 899] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 35) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 900] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 36) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 901] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 37) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 902] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 38) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 903] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 112) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 39) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 1344] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 48) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1345] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 49) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1346] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 50) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1347] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 51) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1348] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 52) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1349] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 53) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1350] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 54) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1351] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 168) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 55) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 1792] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 64) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1793] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 65) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1794] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 66) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1795] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 67) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1796] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 68) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1797] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 69) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1798] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 70) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 1799] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 224) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 71) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 2240] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 80) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2241] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 81) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2242] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 82) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2243] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 83) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2244] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 84) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2245] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 85) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2246] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 86) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2247] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 280) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 87) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 2688] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 96) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2689] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 97) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2690] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 98) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2691] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 99) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2692] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 100) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2693] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 101) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2694] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 102) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 2695] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 336) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 103) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 3136] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 112) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3137] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 113) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3138] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 114) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3139] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 115) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3140] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 116) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3141] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 117) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3142] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 118) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3143] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 392) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 119) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 3584] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 128) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3585] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 129) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3586] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 130) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3587] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 131) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3588] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 132) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3589] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 133) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3590] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 134) % 144]
- kernel_shared_1[threadIdx_x_2 * 8 + 3591] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 448) // 18 * 4608 + cse_var_1 + (threadIdx_x_2 * 8 + 135) % 144]
- with T.launch_thread(threadIdx_x_2, 56):
- kernel_shared_1[threadIdx_x_2 * 8 + 4032] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129024]
- kernel_shared_1[threadIdx_x_2 * 8 + 4033] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129025]
- kernel_shared_1[threadIdx_x_2 * 8 + 4034] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129026]
- kernel_shared_1[threadIdx_x_2 * 8 + 4035] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129027]
- kernel_shared_1[threadIdx_x_2 * 8 + 4036] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129028]
- kernel_shared_1[threadIdx_x_2 * 8 + 4037] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129029]
- kernel_shared_1[threadIdx_x_2 * 8 + 4038] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129030]
- kernel_shared_1[threadIdx_x_2 * 8 + 4039] = kernel_1[blockIdx_x * 147456 + threadIdx_x_2 // 18 * 4608 + cse_var_1 + threadIdx_x_2 % 18 * 8 + 129031]
- with T.launch_thread(threadIdx_x_2, 56):
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4480] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 16]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4481] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 17]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4482] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 18]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4483] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 19]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4484] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 20]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4485] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 21]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4486] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 22]
- if T.likely(threadIdx_x_2 < 16):
- kernel_shared_1[threadIdx_x_2 * 8 + 4487] = kernel_1[blockIdx_x * 147456 + (threadIdx_x_2 + 560) // 18 * 4608 + cse_var_1 + threadIdx_x_2 * 8 + 23]
- for ry_outer_inner, rx_outer_inner in T.grid(3, 3):
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 3] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 4] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 5] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 6] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 3] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 4] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 5] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 6] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 144]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 3] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 4] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 5] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 6] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 288]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 2] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 3] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 4] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 5] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 6] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 432]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 84] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 85] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 86] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 87] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 84] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 85] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 86] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 87] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 153]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 84] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 85] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 86] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 87] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 297]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 81] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 82] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 83] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 84] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 85] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 86] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 87] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 441]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 165] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 166] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 167] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 168] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 18]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 165] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 166] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 167] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 168] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 162]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 165] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 166] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 167] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 168] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 306]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 162] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 163] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 164] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 165] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 166] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 167] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 168] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 450]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 246] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 247] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 248] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 249] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 27]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 246] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 247] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 248] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 249] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 171]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 246] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 247] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 248] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 249] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 315]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 243] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 244] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 245] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 246] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 247] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 248] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 249] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 459]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 324] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 325] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 326] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 327] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 328] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 329] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 330] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 36]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 324] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 325] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 326] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 327] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 328] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 329] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 330] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 180]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 324] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 325] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 326] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 327] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 328] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 329] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 330] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 324]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 324] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 325] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 326] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 327] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 328] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 329] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 330] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 468]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 405] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 406] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 407] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 408] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 409] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 410] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 411] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 45]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 405] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 406] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 407] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 408] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 409] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 410] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 411] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 189]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 405] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 406] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 407] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 408] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 409] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 410] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 411] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 333]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 405] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 406] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 407] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 408] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 409] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 410] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 411] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 477]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 486] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 487] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 488] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 489] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 490] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 491] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 492] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 54]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 486] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 487] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 488] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 489] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 490] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 491] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 492] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 198]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 486] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 487] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 488] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 489] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 490] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 491] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 492] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 342]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 486] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 487] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 488] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 489] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 490] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 491] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 492] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 486]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 567] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 568] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 569] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 570] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 571] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 572] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 573] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 63]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 567] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 568] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 569] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 570] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 571] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 572] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 573] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 207]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 567] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 568] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 569] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 570] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 571] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 572] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 573] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 351]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 567] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 568] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 569] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 570] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 571] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 572] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 573] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 495]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 648] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 649] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 650] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 651] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 652] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 653] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 654] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 72]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 648] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 649] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 650] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 651] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 652] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 653] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 654] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 216]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 648] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 649] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 650] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 651] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 652] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 653] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 654] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 360]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 648] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 649] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 650] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 651] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 652] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 653] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 654] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 504]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 729] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 730] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 731] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 732] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 733] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 734] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 735] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 81]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 729] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 730] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 731] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 732] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 733] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 734] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 735] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 225]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 729] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 730] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 731] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 732] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 733] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 734] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 735] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 369]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 729] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 730] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 731] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 732] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 733] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 734] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 735] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 513]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 810] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 811] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 812] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 813] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 814] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 815] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 816] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 90]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 810] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 811] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 812] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 813] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 814] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 815] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 816] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 234]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 810] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 811] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 812] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 813] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 814] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 815] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 816] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 378]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 810] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 811] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 812] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 813] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 814] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 815] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 816] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 522]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 891] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 892] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 893] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 894] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 895] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 896] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 897] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 99]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 891] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 892] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 893] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 894] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 895] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 896] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 897] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 243]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 891] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 892] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 893] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 894] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 895] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 896] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 897] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 387]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 891] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 892] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 893] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 894] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 895] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 896] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 897] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 531]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 972] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 973] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 974] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 975] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 976] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 977] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 978] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 108]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 972] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 973] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 974] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 975] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 976] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 977] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 978] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 252]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 972] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 973] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 974] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 975] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 976] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 977] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 978] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 396]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 972] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 973] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 974] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 975] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 976] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 977] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 978] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 540]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1053] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1054] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1055] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1056] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1057] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1058] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1059] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 117]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1053] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1054] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1055] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1056] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1057] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1058] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1059] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 261]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1053] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1054] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1055] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1056] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1057] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1058] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1059] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 405]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1053] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1054] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1055] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1056] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1057] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1058] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1059] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 549]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1134] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1135] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1136] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1137] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1138] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1139] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1140] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 126]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1134] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1135] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1136] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1137] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1138] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1139] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1140] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 270]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1134] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1135] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1136] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1137] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1138] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1139] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1140] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 414]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1134] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1135] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1136] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1137] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1138] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1139] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1140] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 558]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1215] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1216] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1217] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1218] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[16] = conv2d_nchw_1[16] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1219] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[20] = conv2d_nchw_1[20] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1220] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[24] = conv2d_nchw_1[24] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1221] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 135]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1215] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1216] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1217] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1218] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[17] = conv2d_nchw_1[17] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1219] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[21] = conv2d_nchw_1[21] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1220] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[25] = conv2d_nchw_1[25] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1221] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 279]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1215] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1216] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1217] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[14] = conv2d_nchw_1[14] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1218] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[18] = conv2d_nchw_1[18] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1219] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[22] = conv2d_nchw_1[22] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1220] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[26] = conv2d_nchw_1[26] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1221] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 423]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1215] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1216] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1217] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[15] = conv2d_nchw_1[15] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1218] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[19] = conv2d_nchw_1[19] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1219] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[23] = conv2d_nchw_1[23] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1220] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- conv2d_nchw_1[27] = conv2d_nchw_1[27] + pad_temp_shared_1[ry_outer_inner * 9 + threadIdx_x % 7 * 9 + rx_outer_inner + 1221] * kernel_shared_1[threadIdx_x // 7 * 576 + ry_outer_inner * 3 + rx_outer_inner + 567]
- for i1_inner in range(4):
+ with T.launch_thread(threadIdx_x_2, 112):
+ kernel_shared_1[threadIdx_x_2 * 3] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 * 3 // 32 * 4608 + cse_var_2 + threadIdx_x_2 * 3 % 32 * 9 + cse_var_1 + rx_outer_outer]
+ kernel_shared_1[threadIdx_x_2 * 3 + 1] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 1) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 1) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ kernel_shared_1[threadIdx_x_2 * 3 + 2] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 2) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 2) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_2, 112):
+ if T.likely(threadIdx_x_2 < 59):
+ kernel_shared_1[threadIdx_x_2 * 3 + 336] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 336) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 16) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ if T.likely(threadIdx_x_2 < 59):
+ kernel_shared_1[threadIdx_x_2 * 3 + 337] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 337) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 17) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ if T.likely(threadIdx_x_2 < 58):
+ kernel_shared_1[threadIdx_x_2 * 3 + 338] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_2 * 3 + 338) // 32 * 4608 + cse_var_2 + (threadIdx_x_2 * 3 + 18) % 32 * 9 + cse_var_1 + rx_outer_outer]
+ for rc_outer_inner in range(16):
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 3] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 4] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 5] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 6] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 49] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 50] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 51] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 52] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 53] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 32 + rc_outer_inner * 2 + 1]
+ for i3_inner in range(7):
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7] = T.max(conv2d_nchw_1[i1_inner] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 1] = T.max(conv2d_nchw_1[i1_inner + 4] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 2] = T.max(conv2d_nchw_1[i1_inner + 8] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 3] = T.max(conv2d_nchw_1[i1_inner + 12] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 4] = T.max(conv2d_nchw_1[i1_inner + 16] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 5] = T.max(conv2d_nchw_1[i1_inner + 20] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
- compute_1[blockIdx_x * 1568 + threadIdx_x // 7 * 196 + i1_inner * 49 + threadIdx_x % 7 * 7 + 6] = T.max(conv2d_nchw_1[i1_inner + 24] + bias_1[blockIdx_x * 32 + threadIdx_x // 7 * 4 + i1_inner], T.float32(0))
+ compute_1[blockIdx_x * 784 + threadIdx_x * 7 + i3_inner] = T.max(conv2d_nchw_1[i3_inner] + bias_1[blockIdx_x * 16 + threadIdx_x // 7], T.float32(0))
</pre></div>
</div>
</div>
@@ -1204,7 +623,7 @@ class Module:
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.265 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.387 ms
</pre></div>
</div>
</div>
@@ -1233,37 +652,37 @@ 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=4)
+conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
-conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=8)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=16)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=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_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=7)
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_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=7)
-conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=16)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=1)
+conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=16)
conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
-conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
+conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
-conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
+conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=4)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=8)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=16)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_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_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
-compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_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)
kernel_shared = s.cache_read(kernel, "shared", [conv2d_nchw])
@@ -1280,16 +699,16 @@ s[compute].bind(compute_i0_o_o_i_i1_o_o_i_fused_i2_o_o_i_fused_i3_o_o_i_fused, t
compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused = s[compute].fuse(compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i)
s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread_axis("threadIdx.x"))
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=8)
+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=3)
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=56)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
-s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 64)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
CUDA source code:
@@ -1307,632 +726,69 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[28];
- __shared__ float pad_temp_shared[1296];
- __shared__ float kernel_shared[4608];
+extern "C" __global__ void __launch_bounds__(112) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[7];
+ __shared__ float pad_temp_shared[1568];
+ __shared__ float kernel_shared[512];
conv2d_nchw[0] = 0.000000e+00f;
- conv2d_nchw[4] = 0.000000e+00f;
- conv2d_nchw[8] = 0.000000e+00f;
- conv2d_nchw[12] = 0.000000e+00f;
- conv2d_nchw[16] = 0.000000e+00f;
- conv2d_nchw[20] = 0.000000e+00f;
- conv2d_nchw[24] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
- conv2d_nchw[5] = 0.000000e+00f;
- conv2d_nchw[9] = 0.000000e+00f;
- conv2d_nchw[13] = 0.000000e+00f;
- conv2d_nchw[17] = 0.000000e+00f;
- conv2d_nchw[21] = 0.000000e+00f;
- conv2d_nchw[25] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
- conv2d_nchw[6] = 0.000000e+00f;
- conv2d_nchw[10] = 0.000000e+00f;
- conv2d_nchw[14] = 0.000000e+00f;
- conv2d_nchw[18] = 0.000000e+00f;
- conv2d_nchw[22] = 0.000000e+00f;
- conv2d_nchw[26] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
- conv2d_nchw[7] = 0.000000e+00f;
- conv2d_nchw[11] = 0.000000e+00f;
- conv2d_nchw[15] = 0.000000e+00f;
- conv2d_nchw[19] = 0.000000e+00f;
- conv2d_nchw[23] = 0.000000e+00f;
- conv2d_nchw[27] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 32; ++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 * 784) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 56)] = (((((9 <= ((((int)threadIdx.x) + 56) % 81)) && (((((int)threadIdx.x) + 56) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 56) / 81) * 49)) + ((((((int)threadIdx.x) + 56) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((9 <= ((((int)threadIdx.x) + 31) % 81)) && (((((int)threadIdx.x) + 31) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 112) / 81) * 49)) + ((((((int)threadIdx.x) + 31) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 168)] = ((((3 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 168) / 81) * 49)) + (((((int)threadIdx.x) + 6) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((9 <= ((((int)threadIdx.x) + 62) % 81)) && (((((int)threadIdx.x) + 62) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 224) / 81) * 49)) + ((((((int)threadIdx.x) + 62) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 280)] = (((((9 <= ((((int)threadIdx.x) + 37) % 81)) && (((((int)threadIdx.x) + 37) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 280) / 81) * 49)) + ((((((int)threadIdx.x) + 37) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 336)] = (((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 336) / 81) * 49)) + (((((int)threadIdx.x) + 12) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 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 * 784) + (((((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) + 448)] = (((((9 <= ((((int)threadIdx.x) + 43) % 81)) && (((((int)threadIdx.x) + 43) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 448) / 81) * 49)) + ((((((int)threadIdx.x) + 43) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 504)] = ((((((int)threadIdx.x) < 54) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 504) / 81) * 49)) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) + 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 560)] = (((((9 <= ((((int)threadIdx.x) + 74) % 81)) && (((((int)threadIdx.x) + 74) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 560) / 81) * 49)) + ((((((int)threadIdx.x) + 74) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 616)] = (((((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 * 784) + (((((int)threadIdx.x) + 616) / 81) * 49)) + ((((((int)threadIdx.x) + 49) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 672)] = ((((((int)threadIdx.x) < 48) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 672) / 81) * 49)) + (((((int)threadIdx.x) + 24) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 728)] = (((((9 <= ((((int)threadIdx.x) + 80) % 81)) && (((((int)threadIdx.x) + 80) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 728) / 81) * 49)) + ((((((int)threadIdx.x) + 80) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((9 <= ((((int)threadIdx.x) + 55) % 81)) && (((((int)threadIdx.x) + 55) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 784) / 81) * 49)) + ((((((int)threadIdx.x) + 55) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 840)] = (((((9 <= ((((int)threadIdx.x) + 30) % 81)) && (((((int)threadIdx.x) + 30) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 840) / 81) * 49)) + ((((((int)threadIdx.x) + 30) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 896)] = ((((4 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 896) / 81) * 49)) + (((((int)threadIdx.x) + 5) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 952)] = (((((9 <= ((((int)threadIdx.x) + 61) % 81)) && (((((int)threadIdx.x) + 61) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 952) / 81) * 49)) + ((((((int)threadIdx.x) + 61) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1008)] = (((((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 * 784) + (((((int)threadIdx.x) + 1008) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 4) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1064)] = (((1 <= ((((int)threadIdx.x) + 2) % 9)) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1064) / 81) * 49)) + (((((int)threadIdx.x) + 11) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1120)] = (((((9 <= ((((int)threadIdx.x) + 67) % 81)) && (((((int)threadIdx.x) + 67) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1120) / 81) * 49)) + ((((((int)threadIdx.x) + 67) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((9 <= ((((int)threadIdx.x) + 42) % 81)) && (((((int)threadIdx.x) + 42) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1176) / 81) * 49)) + ((((((int)threadIdx.x) + 42) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1232)] = ((((((int)threadIdx.x) < 55) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1232) / 81) * 49)) + (((((int)threadIdx.x) + 17) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 8) {
- pad_temp_shared[(((int)threadIdx.x) + 1288)] = 0.000000e+00f;
- }
- kernel_shared[(((int)threadIdx.x) * 8)] = kernel[((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 1)];
- kernel_shared[((((int)threadIdx.x) * 8) + 2)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 2)];
- kernel_shared[((((int)threadIdx.x) * 8) + 3)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 3)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 4)];
- kernel_shared[((((int)threadIdx.x) * 8) + 5)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 5)];
- kernel_shared[((((int)threadIdx.x) * 8) + 6)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 6)];
- kernel_shared[((((int)threadIdx.x) * 8) + 7)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 7)];
- kernel_shared[((((int)threadIdx.x) * 8) + 448)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 16) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 449)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 17) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 450)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 18) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 451)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 19) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 452)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 20) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 453)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 21) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 454)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 22) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 455)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 56) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 23) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 896)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 32) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 897)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 33) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 898)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 34) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 899)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 35) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 900)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 36) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 901)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 37) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 902)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 38) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 903)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 39) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1344)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 48) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1345)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 49) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1346)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 50) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1347)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 51) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1348)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 52) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1349)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 53) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1350)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 54) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1351)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 55) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1792)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 64) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1793)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 65) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1794)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 66) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1795)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 67) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1796)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 68) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1797)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 69) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1798)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 70) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 1799)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 71) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2240)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 80) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2241)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 81) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2242)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 82) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2243)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 83) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2244)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 84) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2245)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 85) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2246)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 86) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2247)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 87) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2688)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 96) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2689)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 97) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2690)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 98) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2691)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 99) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2692)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 100) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2693)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 101) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2694)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 102) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 2695)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 103) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3136)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 112) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3137)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 113) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3138)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 114) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3139)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 115) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3140)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 116) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3141)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 117) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3142)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 118) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3143)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 119) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3584)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 128) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3585)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 129) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3586)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 130) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3587)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 131) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3588)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 132) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3589)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 133) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3590)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 134) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 3591)] = kernel[((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) * 8) + 135) % 144))];
- kernel_shared[((((int)threadIdx.x) * 8) + 4032)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129024)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4033)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129025)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4034)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129026)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4035)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129027)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4036)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129028)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4037)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129029)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4038)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129030)];
- kernel_shared[((((int)threadIdx.x) * 8) + 4039)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) % 18) * 8)) + 129031)];
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4480)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 16)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4481)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 17)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4482)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 18)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4483)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 19)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4484)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 20)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4485)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 21)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4486)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 22)];
- }
- if (((int)threadIdx.x) < 16) {
- kernel_shared[((((int)threadIdx.x) * 8) + 4487)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 8)) + 23)];
- }
- __syncthreads();
- for (int ry_outer_inner = 0; ry_outer_inner < 3; ++ry_outer_inner) {
- for (int rx_outer_inner = 0; rx_outer_inner < 3; ++rx_outer_inner) {
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 3)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 4)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 5)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 6)] * kernel_shared[((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 3)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 4)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 5)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 6)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 144)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 3)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 4)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 5)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 6)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 288)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 3)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 4)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 5)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 6)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 432)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 85)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 86)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 87)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 85)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 86)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 87)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 153)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 85)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 86)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 87)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 297)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 85)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 86)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 87)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 441)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 165)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 166)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 167)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 168)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 165)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 166)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 167)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 168)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 162)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 165)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 166)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 167)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 168)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 306)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 162)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 163)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 164)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 165)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 166)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 167)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 168)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 450)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 246)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 247)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 248)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 249)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 246)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 247)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 248)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 249)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 171)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 246)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 247)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 248)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 249)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 315)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 243)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 244)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 245)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 246)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 247)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 248)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 249)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 459)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 324)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 325)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 326)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 327)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 328)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 329)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 330)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 324)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 325)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 326)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 327)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 328)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 329)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 330)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 180)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 324)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 325)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 326)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 327)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 328)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 329)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 330)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 324)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 324)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 325)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 326)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 327)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 328)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 329)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 330)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 468)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 405)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 406)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 407)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 408)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 409)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 410)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 411)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 405)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 406)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 407)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 408)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 409)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 410)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 411)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 189)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 405)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 406)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 407)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 408)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 409)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 410)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 411)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 333)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 405)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 406)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 407)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 408)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 409)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 410)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 411)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 477)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 486)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 487)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 488)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 489)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 490)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 491)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 492)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 54)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 486)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 487)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 488)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 489)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 490)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 491)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 492)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 198)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 486)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 487)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 488)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 489)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 490)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 491)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 492)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 342)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 486)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 487)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 488)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 489)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 490)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 491)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 492)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 486)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 567)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 568)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 569)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 570)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 571)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 572)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 573)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 63)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 567)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 568)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 569)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 570)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 571)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 572)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 573)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 207)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 567)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 568)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 569)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 570)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 571)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 572)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 573)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 351)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 567)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 568)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 569)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 570)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 571)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 572)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 573)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 495)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 648)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 649)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 650)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 651)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 652)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 653)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 654)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 648)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 649)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 650)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 651)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 652)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 653)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 654)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 216)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 648)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 649)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 650)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 651)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 652)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 653)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 654)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 360)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 648)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 649)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 650)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 651)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 652)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 653)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 654)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 504)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 729)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 730)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 731)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 732)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 733)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 734)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 735)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 81)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 729)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 730)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 731)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 732)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 733)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 734)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 735)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 225)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 729)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 730)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 731)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 732)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 733)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 734)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 735)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 369)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 729)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 730)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 731)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 732)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 733)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 734)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 735)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 513)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 810)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 811)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 812)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 813)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 814)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 815)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 816)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 90)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 810)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 811)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 812)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 813)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 814)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 815)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 816)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 234)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 810)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 811)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 812)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 813)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 814)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 815)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 816)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 378)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 810)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 811)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 812)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 813)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 814)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 815)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 816)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 522)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 891)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 892)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 893)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 894)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 895)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 896)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 897)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 99)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 891)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 892)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 893)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 894)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 895)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 896)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 897)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 243)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 891)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 892)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 893)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 894)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 895)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 896)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 897)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 387)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 891)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 892)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 893)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 894)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 895)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 896)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 897)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 531)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 972)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 973)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 974)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 975)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 976)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 977)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 978)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 108)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 972)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 973)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 974)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 975)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 976)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 977)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 978)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 252)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 972)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 973)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 974)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 975)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 976)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 977)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 978)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 396)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 972)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 973)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 974)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 975)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 976)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 977)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 978)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 540)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1053)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1054)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1055)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1056)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1057)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1058)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1059)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 117)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1053)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1054)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1055)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1056)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1057)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1058)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1059)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 261)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1053)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1054)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1055)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1056)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1057)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1058)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1059)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 405)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1053)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1054)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1055)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1056)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1057)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1058)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1059)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 549)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1134)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1138)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1139)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1140)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 126)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1134)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1138)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1139)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1140)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 270)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1134)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1138)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1139)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1140)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 414)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1134)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1135)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1136)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1137)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1138)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1139)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1140)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 558)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1215)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1219)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1220)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1221)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 135)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1215)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1219)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1220)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1221)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 279)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1215)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1219)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1220)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1221)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 423)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1215)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1216)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1217)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1218)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1219)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1220)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[((((ry_outer_inner * 9) + ((((int)threadIdx.x) % 7) * 9)) + rx_outer_inner) + 1221)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 576) + (ry_outer_inner * 3)) + rx_outer_inner) + 567)]));
+ 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) {
+ for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((((int)threadIdx.x) % 49) / 7) + ry_outer_outer)) && ((((((int)threadIdx.x) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 2) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 2) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 104)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 4) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 4) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 216)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 336)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 6) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 6) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 328)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 1) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 1) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 440)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 560)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 3) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 3) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 552)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 672)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 5) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 5) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 664)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((1 <= (((((int)threadIdx.x) % 49) / 7) + ry_outer_outer)) && ((((((int)threadIdx.x) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 776)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 896)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 2) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 2) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 888)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1008)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 4) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 4) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1000)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1120)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 6) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 6) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1112)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1232)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 1) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 1) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1224)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1344)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 3) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 3) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1336)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1456)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + 5) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + 5) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (ry_outer_outer * 7)) + ((int)threadIdx.x)) + rx_outer_outer) + 1448)] : 0.000000e+00f);
+ kernel_shared[(((int)threadIdx.x) * 3)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) * 3) >> 5) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) * 3) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[((((int)threadIdx.x) * 3) + 1)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 1) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 1) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[((((int)threadIdx.x) * 3) + 2)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 2) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 2) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ if (((int)threadIdx.x) < 59) {
+ kernel_shared[((((int)threadIdx.x) * 3) + 336)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 336) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 16) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ }
+ if (((int)threadIdx.x) < 59) {
+ kernel_shared[((((int)threadIdx.x) * 3) + 337)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 337) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 17) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ }
+ if (((int)threadIdx.x) < 58) {
+ kernel_shared[((((int)threadIdx.x) * 3) + 338)] = kernel[((((((((int)blockIdx.x) * 73728) + ((((((int)threadIdx.x) * 3) + 338) >> 5) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) * 3) + 18) & 31) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ }
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 16; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7))] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2))]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 49)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 50)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 51)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 52)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 53)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 54)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 55)] * kernel_shared[((((((int)threadIdx.x) / 7) * 32) + (rc_outer_inner * 2)) + 1)]));
+ }
}
}
}
- for (int i1_inner = 0; i1_inner < 4; ++i1_inner) {
- compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 1)] = max((conv2d_nchw[(i1_inner + 4)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 2)] = max((conv2d_nchw[(i1_inner + 8)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 3)] = max((conv2d_nchw[(i1_inner + 12)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 4)] = max((conv2d_nchw[(i1_inner + 16)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 5)] = max((conv2d_nchw[(i1_inner + 20)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
- compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 196)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + 6)] = max((conv2d_nchw[(i1_inner + 24)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 4)) + i1_inner)]), 0.000000e+00f);
+ for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
+ compute[(((((int)blockIdx.x) * 784) + (((int)threadIdx.x) * 7)) + i3_inner)] = max((conv2d_nchw[i3_inner] + bias[((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
}
}
</pre></div>
@@ -1967,7 +823,7 @@ 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> ( 6 minutes 25.850 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes 28.831 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 660151b01d..20d2fc96cc 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -921,7 +921,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 8.1379 8.1404 8.1508 8.1224 0.0117
+ 8.1199 8.1161 8.1291 8.1144 0.0065
</pre></div>
</div>
</div>
@@ -943,7 +943,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 10.653 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 13.908 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-network-cuda-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/eafe360d52540634c9eea0fa89e804bd/tune_network_cuda.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_network_cuda.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
index 972cabf050..ac7ba36968 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -940,7 +940,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 757.0472 756.6426 758.7837 755.7152 1.2850
+ 770.8157 771.0289 771.6153 769.8030 0.7551
</pre></div>
</div>
</div>
@@ -962,7 +962,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 42.813 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 47.644 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 d3c69d1162..3495b99212 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -637,27 +637,25 @@ class Module:
@T.prim_func
def main(placeholder: T.Buffer((128, 256), "float32"), placeholder_1: T.Buffer((4916, 16, 1), "float32"), placeholder_2: T.Buffer((4916,), "int32"), placeholder_3: T.Buffer((33,), "int32"), placeholder_4: T.Buffer((128, 512), "float32"), compute: T.Buffer((128, 512), "float32")):
T.func_attr({"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True})
- for i0_outer_i1_outer_fused in T.parallel(512):
- compute_1 = T.allocate([128], "float32", "global")
- compute_2 = T.Buffer((128,), data=compute_1)
- for i_outer_inner in range(2):
+ for i0_outer_i1_outer_fused in T.parallel(32):
+ compute_1 = T.allocate([2048], "float32", "global")
+ compute_2 = T.Buffer((2048,), data=compute_1)
+ for i_outer_inner in range(32):
for i_inner_init, j_init in T.grid(4, 16):
compute_2[i_outer_inner * 64 + i_inner_init * 16 + j_init] = T.float32(0)
- for elem_idx, i_inner, j in T.grid(T.Let(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1], where={cse_var_1: i0_outer_i1_outer_fused % 32}), 4, 16):
- cse_var_1 = T.int32()
+ for elem_idx, i_inner, j in T.grid(placeholder_5[i0_outer_i1_outer_fused + 1] - placeholder_5[i0_outer_i1_outer_fused], 4, 16):
placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_2: T.int32 = i0_outer_i1_outer_fused % 32
- if T.likely(elem_idx < placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2]):
+ if T.likely(elem_idx < placeholder_5[i0_outer_i1_outer_fused + 1] - placeholder_5[i0_outer_i1_outer_fused]):
placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
placeholder_7 = T.Buffer((32768,), data=placeholder.data)
placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
- cse_var_3: T.int32 = i_outer_inner * 64 + i_inner * 16 + j
- compute_2[cse_var_3] = compute_2[cse_var_3] + placeholder_6[placeholder_5[cse_var_2] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 2048 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_2] + elem_idx]], T.float32(0))
- for i0_inner in range(8):
- cse_var_4: T.int32 = i0_outer_i1_outer_fused // 32 * 4096 + i0_inner * 512 + i0_outer_i1_outer_fused % 32 * 16
+ cse_var_1: T.int32 = i_outer_inner * 64 + i_inner * 16 + j
+ compute_2[cse_var_1] = compute_2[cse_var_1] + placeholder_6[placeholder_5[i0_outer_i1_outer_fused] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[i0_outer_i1_outer_fused] + elem_idx]], T.float32(0))
+ for i0_inner in range(128):
+ cse_var_2: T.int32 = i0_inner * 512 + i0_outer_i1_outer_fused * 16
compute_3 = T.Buffer((65536,), data=compute.data)
placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
- compute_3[cse_var_4:cse_var_4 + 16] = T.max(compute_2[i0_inner * 16:i0_inner * 16 + 16] + placeholder_5[cse_var_4:cse_var_4 + 16], T.Broadcast(T.float32(0), 16))
+ compute_3[cse_var_2:cse_var_2 + 16] = T.max(compute_2[i0_inner * 16:i0_inner * 16 + 16] + placeholder_5[cse_var_2:cse_var_2 + 16], T.Broadcast(T.float32(0), 16))
</pre></div>
</div>
</div>
@@ -691,7 +689,7 @@ class Module:
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.361 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.485 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 f86d2673f7..4e01a085dc 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:37.756</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:40.527</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,11 +354,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></td>
-<td><p>00:37.722</p></td>
+<td><p>00:40.491</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>00:00.020</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>
@@ -370,7 +370,7 @@
<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>
-<td><p>00:00.004</p></td>
+<td><p>00:00.005</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
index 0b0a30889f..02eb9c29e0 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -695,8 +695,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 16, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10353238
-No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5478804
+No: 2 GFLOPS: 114.52/114.52 result: MeasureResult(costs=(0.0020214115454545455,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.826885461807251, timestamp=1679626739.1992018) [('tile_f', [-1, 1, 1, 2]), ('tile_y', [-1, 1, 7, 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', 1)],None,6269615
+No: 3 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -818,8 +819,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7754643
-No: 3 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4664196
+No: 4 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -941,8 +942,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 64]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1050261
-No: 4 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 512, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4573854
+No: 5 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1064,9 +1065,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8029103
-No: 5 GFLOPS: 43.22/43.22 result: MeasureResult(costs=(0.005356821111111111,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6050300598144531, timestamp=1679616307.8310463) [('tile_f', [-1, 2, 64, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6649375
-No: 6 GFLOPS: 0.00/43.22 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 64, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4178335
+No: 6 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1188,8 +1188,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6375491
-No: 7 GFLOPS: 0.00/43.22 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7328010
+No: 7 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1311,9 +1311,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8450931
-No: 8 GFLOPS: 0.99/43.22 result: MeasureResult(costs=(0.23476428200000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.164633512496948, timestamp=1679616311.447056) [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5811496
-No: 9 GFLOPS: 0.00/43.22 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9260858
+No: 8 GFLOPS: 2.80/114.52 result: MeasureResult(costs=(0.08256726824999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.456061601638794, timestamp=1679626752.5319996) [('tile_f', [-1, 64, 2, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5821050
+No: 9 GFLOPS: 0.00/114.52 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1435,8 +1435,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9463390
-No: 10 GFLOPS: 0.00/43.22 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2606293
+No: 10 GFLOPS: 155.35/155.35 result: MeasureResult(costs=(0.0014902013766233768,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.0830042362213135, timestamp=1679626755.7926762) [('tile_f', [-1, 8, 4, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4879015
+No: 11 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1558,10 +1559,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,107529
-No: 11 GFLOPS: 201.95/201.95 result: MeasureResult(costs=(0.0011463299279279278,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8168115615844727, timestamp=1679616315.2119982) [('tile_f', [-1, 2, 8, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,6988108
-No: 12 GFLOPS: 2.33/201.95 result: MeasureResult(costs=(0.0994267805,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.338953971862793, timestamp=1679616317.0621138) [('tile_f', [-1, 1, 1, 256]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2328916
-No: 13 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6578462
+No: 12 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1683,8 +1682,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 8, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4705337
-No: 14 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5055302
+No: 13 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1806,8 +1805,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9666307
-No: 15 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 8, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,342501
+No: 14 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1929,8 +1928,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7221425
-No: 16 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5487563
+No: 15 GFLOPS: 0.00/155.35 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2052,8 +2051,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 256, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,901832
-No: 17 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1293394
+No: 16 GFLOPS: 259.48/259.48 result: MeasureResult(costs=(0.0008921655081967214,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.353220224380493, timestamp=1679626758.422771) [('tile_f', [-1, 1, 32, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9729540
+No: 17 GFLOPS: 0.00/259.48 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2175,9 +2175,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 256, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5570672
-No: 18 GFLOPS: 53.19/201.95 result: MeasureResult(costs=(0.0043525985,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2984583377838135, timestamp=1679616319.7499006) [('tile_f', [-1, 1, 1, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7065656
-No: 19 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9016650
+No: 18 GFLOPS: 0.00/259.48 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2299,8 +2298,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 1, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6032098
-No: 20 GFLOPS: 0.00/201.95 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 1, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,53302
+No: 19 GFLOPS: 0.00/259.48 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2422,7 +2421,130 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 64]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7314326
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 1, 128]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,82932
+No: 20 GFLOPS: 0.00/259.48 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
+ func = build(s, args, target=target, runtime=runtime)
+ File "/workspace/python/tvm/driver/build_module.py", line 227, in build
+ input_mod = lower(inputs, args, name=name, binds=binds)
+ File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
+ return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
+ File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+tvm._ffi.base.TVMError: Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc:100
+ 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+ at ../include/tvm/runtime/packed_func.h:1753
+ 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+ at ../include/tvm/runtime/packed_func.h:1697
+ 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ at ../include/tvm/runtime/packed_func.h:1621
+ 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 1: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 0: operator()
+ at ../src/runtime/c_runtime_api.cc:534
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+
+Traceback (most recent call last):
+ 24: TVMFuncCall
+ at ../src/runtime/c_runtime_api.cc:477
+ 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 22: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 21: operator()
+ at ../include/tvm/runtime/packed_func.h:1734
+ 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+ at ../include/tvm/runtime/packed_func.h:1674
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1634
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1649
+ 13: operator()
+ at ../src/driver/driver_api.cc:402
+ 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+ at ../src/driver/driver_api.cc:388
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:283
+ 10: tvm::transform::Pass::operator()(tvm::IRModule) const
+ at ../src/ir/transform.cc:258
+ 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:451
+ 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/ir/transform.cc:274
+ 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+ at ../src/tir/ir/transform.cc:100
+ 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+ at ../include/tvm/runtime/packed_func.h:1753
+ 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+ at ../include/tvm/runtime/packed_func.h:1697
+ 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ at ../include/tvm/runtime/packed_func.h:1621
+ 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 1: Call
+ at ../include/tvm/runtime/packed_func.h:1213
+ 0: operator()
+ at ../src/runtime/c_runtime_api.cc:534
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9460496
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2461,9 +2583,9 @@ and measure running time.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Finish loading 20 records
Best config:
-[('tile_f', [-1, 2, 8, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,6988108
+[('tile_f', [-1, 1, 32, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9729540
Finish loading 20 records
-Time cost of this operator: 0.001576
+Time cost of this operator: 0.001312
</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 3773b53a6c..d3b3a9dea4 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -649,10 +649,10 @@ the tuned operator.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 315.5 98.744 (1, 2, 10, 10, 3) 2 1 [315.5]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.04 0.951 (1, 6, 10, 10) 1 1 [3.04]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.972 0.304 (1, 1, 10, 10, 3) 1 1 [0.972]
-Total_time - 319.512 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 314.2 98.697 (1, 2, 10, 10, 3) 2 1 [314.2]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.151 0.99 (1, 6, 10, 10) 1 1 [3.151]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.996 0.313 (1, 1, 10, 10, 3) 1 1 [0.996]
+Total_time - 318.347 - - - - -
</pre></div>
</div>
</div>
@@ -704,13 +704,13 @@ Total_time -
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 100.2 97.31 (1, 6, 10, 10, 1) 2 1 [100.2]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.801 1.749 (1, 6, 10, 10) 1 1 [1.801]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.968 0.94 (1, 1, 10, 10, 3) 1 1 [0.968]
-Total_time - 102.97 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 136.8 98.078 (1, 6, 10, 10, 1) 2 1 [136.8]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.821 1.305 (1, 6, 10, 10) 1 1 [1.821]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.86 0.617 (1, 3, 10, 10, 1) 1 1 [0.86]
+Total_time - 139.481 - - - - -
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 20.905 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 28.427 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/9ccca8fd489a1486ac71b55a55c320c5/micro_autotune.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_autotune.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_pytorch.html b/docs/how_to/work_with_microtvm/micro_pytorch.html
index 381a26cd0b..921d0ae5c3 100644
--- a/docs/how_to/work_with_microtvm/micro_pytorch.html
+++ b/docs/how_to/work_with_microtvm/micro_pytorch.html
@@ -460,8 +460,8 @@ download a cat image and preprocess it to use as the model input.</p>
Downloading: "https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
0%| | 0.00/3.42M [00:00<?, ?B/s]
- 61%|###### | 2.09M/3.42M [00:00<00:00, 17.9MB/s]
-100%|##########| 3.42M/3.42M [00:00<00:00, 27.9MB/s]
+ 61%|###### | 2.09M/3.42M [00:00<00:00, 16.8MB/s]
+100%|##########| 3.42M/3.42M [00:00<00:00, 26.4MB/s]
/workspace/python/tvm/relay/frontend/pytorch_utils.py:47: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
return LooseVersion(torch_ver) > ver
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
@@ -587,7 +587,7 @@ via the host <cite>main.cc`</cite> or if a Zephyr emulated board is selected as
Torch top-1 id: 282, class name: tiger cat
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 16.988 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 22.970 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-pytorch-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/12b9ecc04c41abaa12022061771821d1/micro_pytorch.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_pytorch.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index 51431fde89..b5a360f158 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -528,7 +528,7 @@ take about <strong>2 minutes</strong> to download the Stanford Cars, while COCO
<a href="https://docs.python.org/3/library/shutil.html#shutil.move" title="shutil.move" class="sphx-glr-backref-module-shutil sphx-glr-backref-type-py-function"><span class="n">shutil</span><span class="o">.</span><span class="n">move</span></a><span class="p">(</span><span class="sa">f</span><span class="s2">"</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-typ [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmpsq8t5s21/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmpr8om6t_s/images/random'
</pre></div>
</div>
</div>
@@ -588,8 +588,8 @@ objects to other stuff? We can display some examples from our datasets using <co
<span class="n">plt</span><span class="o">.</span><span class="n">axis</span><span class="p">(</span><span class="s2">"off"</span><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpsq8t5s21/images/target contains 8144 images
-/tmp/tmpsq8t5s21/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpr8om6t_s/images/target contains 8144 images
+/tmp/tmpr8om6t_s/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -701,13 +701,13 @@ the time on our validation set).</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Epoch 1/3
-328/328 - 41s - loss: 0.2322 - accuracy: 0.9176 - val_loss: 0.1163 - val_accuracy: 0.9558 - 41s/epoch - 124ms/step
+328/328 - 42s - loss: 0.2768 - accuracy: 0.9157 - val_loss: 0.1518 - val_accuracy: 0.9517 - 42s/epoch - 128ms/step
Epoch 2/3
-328/328 - 35s - loss: 0.0927 - accuracy: 0.9660 - val_loss: 0.0900 - val_accuracy: 0.9656 - 35s/epoch - 106ms/step
+328/328 - 35s - loss: 0.1149 - accuracy: 0.9561 - val_loss: 0.1394 - val_accuracy: 0.9513 - 35s/epoch - 108ms/step
Epoch 3/3
-328/328 - 35s - loss: 0.0644 - accuracy: 0.9756 - val_loss: 0.1403 - val_accuracy: 0.9468 - 35s/epoch - 106ms/step
+328/328 - 35s - loss: 0.0679 - accuracy: 0.9748 - val_loss: 0.1042 - val_accuracy: 0.9653 - 35s/epoch - 107ms/step
-<keras.callbacks.History object at 0x7fc54d3a6890>
+<keras.callbacks.History object at 0x7f228f798990>
</pre></div>
</div>
</div>
@@ -971,7 +971,7 @@ as intended.</p>
<p>From here, we could modify the model to read live images from the camera - we have another
Arduino tutorial for how to do that <a class="reference external" href="https://github.com/guberti/tvm-arduino-demos/tree/master/examples/person_detection">on GitHub</a>. Alternatively, we could also
<a class="reference external" href="https://tvm.apache.org/docs/how_to/work_with_microtvm/micro_autotune.html">use TVM’s autotuning capabilities</a> to dramatically improve the model’s performance.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 22.921 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 31.483 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 feb2389729..45eb7a3ebe 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>07:26.229</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>07:50.419</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 82%" />
@@ -354,27 +354,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_train.html#sphx-glr-how-to-work-with-microtvm-micro-train-py"><span class="std std-ref">5. Training Vision Models for microTVM on Arduino</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_train.py</span></code>)</p></td>
-<td><p>04:22.921</p></td>
+<td><p>04:31.483</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">6. Model Tuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>01:20.905</p></td>
+<td><p>01:28.427</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_pytorch.html#sphx-glr-how-to-work-with-microtvm-micro-pytorch-py"><span class="std std-ref">4. microTVM PyTorch Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_pytorch.py</span></code>)</p></td>
-<td><p>01:16.988</p></td>
+<td><p>01:22.970</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_aot.html#sphx-glr-how-to-work-with-microtvm-micro-aot-py"><span class="std std-ref">3. microTVM Ahead-of-Time (AOT) Compilation</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_aot.py</span></code>)</p></td>
-<td><p>00:10.223</p></td>
+<td><p>00:10.957</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_custom_ide.html#sphx-glr-how-to-work-with-microtvm-micro-custom-ide-py"><span class="std std-ref">9. Bring microTVM to your own development environment</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_custom_ide.py</span></code>)</p></td>
-<td><p>00:07.873</p></td>
+<td><p>00:08.651</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">2. microTVM TFLite Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:07.318</p></td>
+<td><p>00:07.931</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">7. Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index 4ad3005b35..c0156e2dc6 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-relay-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:36.852</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:38.525</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,15 +354,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="using_pipeline_executor.html#sphx-glr-how-to-work-with-relay-using-pipeline-executor-py"><span class="std std-ref">Using Pipeline Executor in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_pipeline_executor.py</span></code>)</p></td>
-<td><p>00:32.206</p></td>
+<td><p>00:33.774</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></td>
-<td><p>00:02.895</p></td>
+<td><p>00:03.004</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></td>
-<td><p>00:01.744</p></td>
+<td><p>00:01.740</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_schedules/intrin_math.html b/docs/how_to/work_with_schedules/intrin_math.html
index 17c2fc5e24..234e64ec7b 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -540,7 +540,7 @@ 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 0x7fc19cc05d40>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7f1edd6a7710>
</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 77ff8a4cd2..934baadb32 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:07.653</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:08.721</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,35 +354,35 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></td>
-<td><p>00:04.907</p></td>
+<td><p>00:05.745</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></td>
-<td><p>00:01.241</p></td>
+<td><p>00:01.410</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>00:00.620</p></td>
+<td><p>00:00.642</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>00:00.604</p></td>
+<td><p>00:00.631</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>00:00.133</p></td>
+<td><p>00:00.138</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></td>
-<td><p>00:00.062</p></td>
+<td><p>00:00.066</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></td>
-<td><p>00:00.056</p></td>
+<td><p>00:00.058</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></td>
-<td><p>00:00.030</p></td>
+<td><p>00:00.032</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/install/nnpack.html b/docs/install/nnpack.html
index 1515c4b747..e285b427a5 100644
--- a/docs/install/nnpack.html
+++ b/docs/install/nnpack.html
@@ -234,7 +234,17 @@
<p class="caption" role="heading"><span class="caption-text">Getting Started</span></p>
<ul class="current">
<li class="toctree-l1 current"><a class="reference internal" href="index.html">Installing TVM</a><ul class="current">
-<li class="toctree-l2"><a class="reference internal" href="from_source.html">Install from Source</a></li>
+<li class="toctree-l2 current"><a class="reference internal" href="from_source.html">Install from Source</a><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="from_source.html#developers-get-source-from-github">Developers: Get Source from Github</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_source.html#build-the-shared-library">Build the Shared Library</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_source.html#python-package-installation">Python Package Installation</a></li>
+<li class="toctree-l3 current"><a class="reference internal" href="from_source.html#install-contrib-libraries">Install Contrib Libraries</a><ul class="current">
+<li class="toctree-l4 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a></li>
+</ul>
+</li>
+<li class="toctree-l3"><a class="reference internal" href="from_source.html#enable-c-tests">Enable C++ Tests</a></li>
+</ul>
+</li>
<li class="toctree-l2"><a class="reference internal" href="docker.html">Docker Images</a></li>
<li class="toctree-l2 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#conditions">Conditions</a></li>
diff --git a/docs/reference/api/doxygen/arg__info_8h_source.html b/docs/reference/api/doxygen/arg__info_8h_source.html
index 0bf74745d2..82476e83c1 100644
--- a/docs/reference/api/doxygen/arg__info_8h_source.html
+++ b/docs/reference/api/doxygen/arg__info_8h_source.html
@@ -81,7 +81,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1meta__schedule_1_1TensorInfo_html"><div class="ttname"><a href="classtvm_1_1meta__schedule_1_1TensorInfo.html">tvm::meta_schedule::TensorInfo</a></div><div class="ttdoc">Managed reference to TensorInfoNode. </div><div class="ttdef"><b>Definition:</b> arg_info.h:102</div></div>
<div class="ttc" id="classtvm_1_1meta__schedule_1_1TensorInfoNode_html"><div class="ttname"><a href="classtvm_1_1meta__schedule_1_1TensorInfoNode.html">tvm::meta_schedule::TensorInfoNode</a></div><div class="ttdoc">The tensor argument information. </div><div class="ttdef"><b>Definition:</b> arg_info.h:79</div></div>
<div class="ttc" id="classtvm_1_1meta__schedule_1_1TensorInfoNode_html_ab54cd62d85a051b78c9605ac0d2d3cb7"><div class="ttname"><a href="classtvm_1_1meta__schedule_1_1TensorInfoNode.html#ab54cd62d85a051b78c9605ac0d2d3cb7">tvm::meta_schedule::TensorInfoNode::VisitAttrs</a></div><div class="ttdeci">void VisitAttrs(tvm::AttrVisitor *v)</div><div class="ttdef"><b>Definition:</b> arg_info.h:86</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ObjectRef_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></div><div class="ttdoc">Base class of all object reference. </div><div class="ttdef"><b>Definition:</b> object.h:511</div></div>
<div class="ttc" id="classtvm_1_1meta__schedule_1_1TensorInfoNode_html_a703cc684f922c4c7dddb20a503a78eb9"><div class="ttname"><a href="classtvm_1_1meta__schedule_1_1TensorInfoNode.html#a703cc684f922c4c7dddb20a503a78eb9">tvm::meta_schedule::TensorInfoNode::dtype</a></div><div class="ttdeci">runtime::DataType dtype</div><div class="ttdoc">The data type of the tensor. </div><div class="ttdef"><b>Definition:</b> arg_info.h:82</div></div>
<div class="ttc" id="shape__tuple_8h_html"><div class="ttname"><a href="shape__tuple_8h.html">shape_tuple.h</a></div><div class="ttdoc">Runtime ShapeTuple container types. </div></div>
diff --git a/docs/reference/api/doxygen/bytecode_8h_source.html b/docs/reference/api/doxygen/bytecode_8h_source.html
index 546e823b38..a1fac4106f 100644
--- a/docs/reference/api/doxygen/bytecode_8h_source.html
+++ b/docs/reference/api/doxygen/bytecode_8h_source.html
@@ -111,7 +111,7 @@ $(function() {
<div class="ttc" id="structtvm_1_1runtime_1_1vm_1_1Instruction_html_a46879dbe84105fb621a6167f8d73b223"><div class="ttname"><a href="structtvm_1_1runtime_1_1vm_1_1Instruction.html#a46879dbe84105fb621a6167f8d73b223">tvm::runtime::vm::Instruction::target</a></div><div class="ttdeci">RegName target</div><div class="ttdoc">The register containing the target value. </div><div class="ttdef"><b>Definition:</b> bytecode.h:143</div></div>
<div class="ttc" id="structtvm_1_1runtime_1_1vm_1_1Instruction_html_a360b264ed892e620935b648e5a91a5ea"><div class="ttname"><a href="structtvm_1_1runtime_1_1vm_1_1Instruction.html#a360b264ed892e620935b648e5a91a5ea">tvm::runtime::vm::Instruction::arity</a></div><div class="ttdeci">Index arity</div><div class="ttdoc">The arity of the packed function. </div><div class="ttdef"><b>Definition:</b> bytecode.h:133</div></div>
<div class="ttc" id="structtvm_1_1runtime_1_1vm_1_1Instruction_html_a3a175836bc0893d99935f32911e45bfd"><div class="ttname"><a href="structtvm_1_1runtime_1_1vm_1_1Instruction.html#a3a175836bc0893d99935f32911e45bfd">tvm::runtime::vm::Instruction::closure</a></div><div class="ttdeci">RegName closure</div><div class="ttdoc">The register containing the closure. </div><div class="ttdef"><b>Definition:</b> bytecode.h:115</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="structtvm_1_1runtime_1_1vm_1_1Instruction_html_a3412cabd3b4f42f106f56fc22257f6ca"><div class="ttname"><a href="structtvm_1_1runtime_1_1vm_1_1Instruction.html#a3412cabd3b4f42f106f56fc22257f6ca">tvm::runtime::vm::Instruction::storage</a></div><div class="ttdeci">RegName storage</div><div class="ttdoc">The storage to allocate from. </div><div class="ttdef"><b>Definition:</b> bytecode.h:93</div></div>
<div class="ttc" id="namespacetvm_1_1runtime_1_1vm_html_a3597867d2db714bf760876a23d6b7d3d"><div class="ttname"><a href="namespacetvm_1_1runtime_1_1vm.html#a3597867d2db714bf760876a23d6b7d3d">tvm::runtime::vm::Index</a></div><div class="ttdeci">int64_t Index</div><div class="ttdoc">An alias for the integer type used ubiquitously in the VM. </div><div class="ttdef"><b>Definition:</b> bytecode.h:43</div></div>
<div class="ttc" id="structtvm_1_1runtime_1_1vm_1_1Instruction_html_a5d98a3b5b50e6fa9029d05f741c0bce9"><div class="ttname"><a href="structtvm_1_1runtime_1_1vm_1_1Instruction.html#a5d98a3b5b50e6fa9029d05f741c0bce9">tvm::runtime::vm::Instruction::num_args</a></div><div class="ttdeci">Index num_args</div><div class="ttdoc">The number of arguments to the function. </div><div class="ttdef"><b>Definition:</b> bytecode.h:153</div></div>
diff --git a/docs/reference/api/doxygen/c__runtime__api_8h_source.html b/docs/reference/api/doxygen/c__runtime__api_8h_source.html
index 1dc1b4d287..b64487542b 100644
--- a/docs/reference/api/doxygen/c__runtime__api_8h_source.html
+++ b/docs/reference/api/doxygen/c__runtime__api_8h_source.html
@@ -125,7 +125,7 @@ $(function() {
<div class="ttc" id="c__runtime__api_8h_html_ae690840d1af9c7b0fe5b9b457456f60d"><div class="ttname"><a href="c__runtime__api_8h.html#ae690840d1af9c7b0fe5b9b457456f60d">TVMArrayToDLPack</a></div><div class="ttdeci">int TVMArrayToDLPack(TVMArrayHandle from, DLManagedTensor **out)</div><div class="ttdoc">Produce a DLMangedTensor from the array that shares data memory with the array. </div></div>
<div class="ttc" id="c__runtime__api_8h_html_a6cd1076476117e74454f67931c2da1d4"><div class="ttname"><a href="c__runtime__api_8h.html#a6cd1076476117e74454f67931c2da1d4">TVMRetValueHandle</a></div><div class="ttdeci">void * TVMRetValueHandle</div><div class="ttdoc">Handle to hold return value. </div><div class="ttdef"><b>Definition:</b> c_runtime_api.h:231</div></div>
<div class="ttc" id="c__runtime__api_8h_html_acf57d257a6e0841d84ebbd2a339d183e"><div class="ttname"><a href="c__runtime__api_8h.html#acf57d257a6e0841d84ebbd2a339d183e">TVMFunctionHandle</a></div><div class="ttdeci">void * TVMFunctionHandle</div><div class="ttdoc">Handle to packed function handle. </div><div class="ttdef"><b>Definition:</b> c_runtime_api.h:229</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="c__runtime__api_8h_html_ace8007daffd9f2c6d954c24d870bfcc4"><div class="ttname"><a href="c__runtime__api_8h.html#ace8007daffd9f2c6d954c24d870bfcc4">tvm_index_t</a></div><div class="ttdeci">int64_t tvm_index_t</div><div class="ttdoc">type of array index. </div><div class="ttdef"><b>Definition:</b> c_runtime_api.h:88</div></div>
<div class="ttc" id="c__runtime__api_8h_html_a190e81769e805cca153514137a66e793a9387f774bc8453afe4aa4cd17789a405"><div class="ttname"><a href="c__runtime__api_8h.html#a190e81769e805cca153514137a66e793a9387f774bc8453afe4aa4cd17789a405">kTVMOpaqueHandle</a></div><div class="ttdef"><b>Definition:</b> c_runtime_api.h:177</div></div>
<div class="ttc" id="c__runtime__api_8h_html_ae899a6a211b7913e92420a01b804db64"><div class="ttname"><a href="c__runtime__api_8h.html#ae899a6a211b7913e92420a01b804db64">TVMObjectRetain</a></div><div class="ttdeci">int TVMObjectRetain(TVMObjectHandle obj)</div><div class="ttdoc">Increase the reference count of an object. </div></div>
diff --git a/docs/reference/api/doxygen/data__layout_8h_source.html b/docs/reference/api/doxygen/data__layout_8h_source.html
index d86a9b9fc0..92e6db0da5 100644
--- a/docs/reference/api/doxygen/data__layout_8h_source.html
+++ b/docs/reference/api/doxygen/data__layout_8h_source.html
@@ -96,7 +96,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html">tvm::runtime::Array</a></div><div class="ttdoc">Array, container representing a contiguous sequence of ObjectRefs. </div><div class="ttdef"><b>Definition:</b> array.h:289</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1String_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1String.html">tvm::runtime::String</a></div><div class="ttdoc">Reference to string objects. </div><div class="ttdef"><b>Definition:</b> string.h:98</div></div>
<div class="ttc" id="classtvm_1_1tir_1_1Layout_html_a7fd4348d39ad4249daf89d54195fa65a"><div class="ttname"><a href="classtvm_1_1tir_1_1Layout.html#a7fd4348d39ad4249daf89d54195fa65a">tvm::tir::Layout::ndim</a></div><div class="ttdeci">size_t ndim() const</div><div class="ttdef"><b>Definition:</b> data_layout.h:180</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="object_8h_html_ac6e7295a4999e2c8e4a2c990beca887a"><div class="ttname"><a href="object_8h.html#ac6e7295a4999e2c8e4a2c990beca887a">TVM_DEFINE_OBJECT_REF_METHODS</a></div><div class="ttdeci">#define TVM_DEFINE_OBJECT_REF_METHODS(TypeName, ParentType, ObjectName)</div><div class="ttdef"><b>Definition:</b> object.h:713</div></div>
<div class="ttc" id="classtvm_1_1tir_1_1BijectiveLayoutNode_html"><div class="ttname"><a href="classtvm_1_1tir_1_1BijectiveLayoutNode.html">tvm::tir::BijectiveLayoutNode</a></div><div class="ttdef"><b>Definition:</b> data_layout.h:295</div></div>
<div class="ttc" id="namespacetvm_1_1te_html_ae0c71f84710b436cbe0b32289d0838f4"><div class="ttname"><a href="namespacetvm_1_1te.html#ae0c71f84710b436cbe0b32289d0838f4">tvm::te::var</a></div><div class="ttdeci">Var var(std::string name_hint, DataType t=DataType::Int(32))</div><div class="ttdoc">Construct a new Var expression. </div></div>
diff --git a/docs/reference/api/doxygen/dataflow__pattern_8h_source.html b/docs/reference/api/doxygen/dataflow__pattern_8h_source.html
index 539acf2237..a7e6640a80 100644
--- a/docs/reference/api/doxygen/dataflow__pattern_8h_source.html
+++ b/docs/reference/api/doxygen/dataflow__pattern_8h_source.html
@@ -128,7 +128,7 @@ $(function() {
<div class="ttc" id="namespacetvm_1_1relay_html_a3834130d66634d032f8e9bffafbe09ad"><div class="ttname"><a href="namespacetvm_1_1relay.html#a3834130d66634d032f8e9bffafbe09ad">tvm::relay::IsTuple</a></div><div class="ttdeci">DFPattern IsTuple(const Array< DFPattern > &fields)</div><div class="ttdoc">Syntatic Sugar for creating a TuplePattern. </div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1String_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1String.html">tvm::runtime::String</a></div><div class="ttdoc">Reference to string objects. </div><div class="ttdef"><b>Definition:</b> string.h:98</div></div>
<div class="ttc" id="classtvm_1_1relay_1_1FunctionPatternNode_html_a5e831242707df5d59474816c416abe97"><div class="ttname"><a href="classtvm_1_1relay_1_1FunctionPatternNode.html#a5e831242707df5d59474816c416abe97">tvm::relay::FunctionPatternNode::VisitAttrs</a></div><div class="ttdeci">void VisitAttrs(tvm::AttrVisitor *v)</div><div class="ttdef"><b>Definition:</b> dataflow_pattern.h:201</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="classtvm_1_1RelayExpr_html"><div class="ttname"><a href="classtvm_1_1RelayExpr.html">tvm::RelayExpr</a></div><div class="ttdoc">Managed reference to RelayExprNode. </div><div class="ttdef"><b>Definition:</b> expr.h:433</div></div>
<div class="ttc" id="classtvm_1_1relay_1_1ConstantPatternNode_html"><div class="ttname"><a href="classtvm_1_1relay_1_1ConstantPatternNode.html">tvm::relay::ConstantPatternNode</a></div><div class="ttdoc">Container for Constant. </div><div class="ttdef"><b>Definition:</b> dataflow_pattern.h:138</div></div>
<div class="ttc" id="object_8h_html_ac6e7295a4999e2c8e4a2c990beca887a"><div class="ttname"><a href="object_8h.html#ac6e7295a4999e2c8e4a2c990beca887a">TVM_DEFINE_OBJECT_REF_METHODS</a></div><div class="ttdeci">#define TVM_DEFINE_OBJECT_REF_METHODS(TypeName, ParentType, ObjectName)</div><div class="ttdef"><b>Definition:</b> object.h:713</div></div>
diff --git a/docs/reference/api/doxygen/detail_2broadcast_8h_source.html b/docs/reference/api/doxygen/detail_2broadcast_8h_source.html
index 6122be4cee..2b36a0d807 100644
--- a/docs/reference/api/doxygen/detail_2broadcast_8h_source.html
+++ b/docs/reference/api/doxygen/detail_2broadcast_8h_source.html
@@ -77,7 +77,7 @@ $(function() {
<div class="ttc" id="namespacetvm_html_a0df5ca82d2c566f628ebb2f1e84a3fcb"><div class="ttname"><a href="namespacetvm.html#a0df5ca82d2c566f628ebb2f1e84a3fcb">tvm::max</a></div><div class="ttdeci">PrimExpr max(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">take maximum of two values </div></div>
<div class="ttc" id="namespacetvm_1_1tir_html_ae8c7db788e840dc1c2ed1f365d5ea829"><div class="ttname"><a href="namespacetvm_1_1tir.html#ae8c7db788e840dc1c2ed1f365d5ea829">tvm::tir::IntImmNode</a></div><div class="ttdeci">tvm::IntImmNode IntImmNode</div><div class="ttdef"><b>Definition:</b> expr.h:49</div></div>
<div class="ttc" id="namespacetvm_1_1tir_html_aed3f57cf8d1c3546f075701898c5b70f"><div class="ttname"><a href="namespacetvm_1_1tir.html#aed3f57cf8d1c3546f075701898c5b70f">tvm::tir::make_zero</a></div><div class="ttdeci">PrimExpr make_zero(DataType t, Span span=Span())</div><div class="ttdoc">Make a const zero expr. </div><div class="ttdef"><b>Definition:</b> op.h:962</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html_a6b097149e69ea03fe3b812a3f5f7fcd9"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html#a6b097149e69ea03fe3b812a3f5f7fcd9">tvm::runtime::Array::end</a></div><div class="ttdeci">iterator end() const</div><div class="ttdef"><b>Definition:</b> array.h:390</div></div>
<div class="ttc" id="classtvm_1_1te_1_1Tensor_html"><div class="ttname"><a href="classtvm_1_1te_1_1Tensor.html">tvm::te::Tensor</a></div><div class="ttdoc">Tensor structure representing a possible input, or intermediate computation result. </div><div class="ttdef"><b>Definition:</b> tensor.h:102</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html_ae9734c8e6324f9be27fce8d45d0aa6f4"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html#ae9734c8e6324f9be27fce8d45d0aa6f4">tvm::runtime::Array::begin</a></div><div class="ttdeci">iterator begin() const</div><div class="ttdef"><b>Definition:</b> array.h:387</div></div>
diff --git a/docs/reference/api/doxygen/detail_2extern_8h_source.html b/docs/reference/api/doxygen/detail_2extern_8h_source.html
index 1c16be735d..d8d1d07b57 100644
--- a/docs/reference/api/doxygen/detail_2extern_8h_source.html
+++ b/docs/reference/api/doxygen/detail_2extern_8h_source.html
@@ -74,7 +74,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1runtime_1_1DataType_html_aebad9f7235dd20af649fb5c2113797b8"><div class="ttname"><a href="classtvm_1_1runtime_1_1DataType.html#aebad9f7235dd20af649fb5c2113797b8">tvm::runtime::DataType::Handle</a></div><div class="ttdeci">static DataType Handle(int bits=64, int lanes=1)</div><div class="ttdoc">Construct a handle type. </div><div class="ttdef"><b>Definition:</b> data_type.h:198</div></div>
<div class="ttc" id="namespacetvm_1_1tir_1_1builtin_html_a28f99e6dd767482765b854ee9fc71f2c"><div class="ttname"><a href="namespacetvm_1_1tir_1_1builtin.html#a28f99e6dd767482765b854ee9fc71f2c">tvm::tir::builtin::tvm_stack_make_array</a></div><div class="ttdeci">const Op & tvm_stack_make_array()</div><div class="ttdoc">Allocate a NDArray(DLTensor) on stack, return the handle. </div></div>
<div class="ttc" id="classtvm_1_1te_1_1ExternOp_html"><div class="ttname"><a href="classtvm_1_1te_1_1ExternOp.html">tvm::te::ExternOp</a></div><div class="ttdoc">Managed reference to ExternOpNode. </div><div class="ttdef"><b>Definition:</b> operation.h:460</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="namespacetvm_1_1tir_1_1builtin_html_abd540cb73407771ecfb4f78722ce5a1b"><div class="ttname"><a href="namespacetvm_1_1tir_1_1builtin.html#abd540cb73407771ecfb4f78722ce5a1b">tvm::tir::builtin::tvm_stack_make_shape</a></div><div class="ttdeci">const Op & tvm_stack_make_shape()</div><div class="ttdoc">Allocate a shape tuple on stack, return the handle. </div></div>
<div class="ttc" id="namespacetvm_1_1te_html_ae0c71f84710b436cbe0b32289d0838f4"><div class="ttname"><a href="namespacetvm_1_1te.html#ae0c71f84710b436cbe0b32289d0838f4">tvm::te::var</a></div><div class="ttdeci">Var var(std::string name_hint, DataType t=DataType::Int(32))</div><div class="ttdoc">Construct a new Var expression. </div></div>
<div class="ttc" id="operation_8h_html"><div class="ttname"><a href="operation_8h.html">operation.h</a></div><div class="ttdoc">Operation node can generate one or multiple Tensors. </div></div>
diff --git a/docs/reference/api/doxygen/device__api_8h_source.html b/docs/reference/api/doxygen/device__api_8h_source.html
index b81787d236..dc98268de5 100644
--- a/docs/reference/api/doxygen/device__api_8h_source.html
+++ b/docs/reference/api/doxygen/device__api_8h_source.html
@@ -94,7 +94,7 @@ $(function() {
<div class="ttc" id="namespacetvm_1_1runtime_html_a46fef1ca0ccc05473e9bb0a8c6b66619adff7742765a9f6f50973675bf34ad264"><div class="ttname"><a href="namespacetvm_1_1runtime.html#a46fef1ca0ccc05473e9bb0a8c6b66619adff7742765a9f6f50973675bf34ad264">tvm::runtime::kMaxSharedMemoryPerBlock</a></div><div class="ttdef"><b>Definition:</b> device_api.h:42</div></div>
<div class="ttc" id="namespacetvm_1_1runtime_html_a46fef1ca0ccc05473e9bb0a8c6b66619a90ebfaf325917db841553c65ce2ae630"><div class="ttname"><a href="namespacetvm_1_1runtime.html#a46fef1ca0ccc05473e9bb0a8c6b66619a90ebfaf325917db841553c65ce2ae630">tvm::runtime::kMaxClockRate</a></div><div class="ttdef"><b>Definition:</b> device_api.h:45</div></div>
<div class="ttc" id="namespacetvm_1_1runtime_html_a46fef1ca0ccc05473e9bb0a8c6b66619a100cda550bad77fbaf993566c022bdf9"><div class="ttname"><a href="namespacetvm_1_1runtime.html#a46fef1ca0ccc05473e9bb0a8c6b66619a100cda550bad77fbaf993566c022bdf9">tvm::runtime::kDriverVersion</a></div><div class="ttdef"><b>Definition:</b> device_api.h:51</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="namespacetvm_html_a7c2095aed90b2129ba631b90103313a2"><div class="ttname"><a href="namespacetvm.html#a7c2095aed90b2129ba631b90103313a2">tvm::Device</a></div><div class="ttdeci">DLDevice Device</div><div class="ttdef"><b>Definition:</b> ndarray.h:43</div></div>
<div class="ttc" id="namespacetvm_1_1runtime_html_a46fef1ca0ccc05473e9bb0a8c6b66619aa707ea3fd97e625364ad60daa89be2da"><div class="ttname"><a href="namespacetvm_1_1runtime.html#a46fef1ca0ccc05473e9bb0a8c6b66619aa707ea3fd97e625364ad60daa89be2da">tvm::runtime::kWarpSize</a></div><div class="ttdef"><b>Definition:</b> device_api.h:41</div></div>
<div class="ttc" id="namespacetvm_1_1runtime_html_a46fef1ca0ccc05473e9bb0a8c6b66619ad7ca96c2095b4670de00ee1d0ca79201"><div class="ttname"><a href="namespacetvm_1_1runtime.html#a46fef1ca0ccc05473e9bb0a8c6b66619ad7ca96c2095b4670de00ee1d0ca79201">tvm::runtime::kGcnArch</a></div><div class="ttdef"><b>Definition:</b> device_api.h:49</div></div>
diff --git a/docs/reference/api/doxygen/elemwise_8h_source.html b/docs/reference/api/doxygen/elemwise_8h_source.html
index 77d80b7452..56c2ce94f2 100644
--- a/docs/reference/api/doxygen/elemwise_8h_source.html
+++ b/docs/reference/api/doxygen/elemwise_8h_source.html
@@ -109,7 +109,7 @@ $(function() {
<div class="ttc" id="namespacetvm_1_1topi_html_af9694f5470ba2cabc19866be3b00fe8d"><div class="ttname"><a href="namespacetvm_1_1topi.html#af9694f5470ba2cabc19866be3b00fe8d">tvm::topi::sinh</a></div><div class="ttdeci">Tensor sinh(const Tensor &x, std::string name="T_" "sinh", std::string tag=kElementWise)</div><div class="ttdef"><b>Definition:</b> elemwise.h:66</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a99f3fd2edfd74113c83bc9fb98e8aca1"><div class="ttname"><a href="namespacetvm_1_1topi.html#a99f3fd2edfd74113c83bc9fb98e8aca1">tvm::topi::bitwise_not</a></div><div class="ttdeci">Tensor bitwise_not(const Tensor &x, std::string name="T_bitwise_not", std::string tag=kElementWise)</div><div class="ttdoc">Creates an operation that returns the bitwise NOT of a given tensor. </div><div class="ttdef"><b>Definition:</b> elemwise.h:197</d [...]
<div class="ttc" id="namespacetvm_1_1topi_html_aec153e599d33c78a7592007cde1c02cb"><div class="ttname"><a href="namespacetvm_1_1topi.html#aec153e599d33c78a7592007cde1c02cb">tvm::topi::tanh</a></div><div class="ttdeci">Tensor tanh(const Tensor &x, std::string name="T_" "tanh", std::string tag=kElementWise)</div><div class="ttdef"><b>Definition:</b> elemwise.h:74</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a188f69f0e65e2efb4914e458db50b3d8"><div class="ttname"><a href="namespacetvm_1_1topi.html#a188f69f0e65e2efb4914e458db50b3d8">tvm::topi::fast_erf_float32</a></div><div class="ttdeci">Tensor fast_erf_float32(const Tensor &data, std::string name, std::string tag)</div><div class="ttdoc">Fast_erf_float expression from Eigen. </div><div class="ttdef"><b>Definition:</b> elemwise.h:462</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a4435c7d12111817c485ebe9d2a6d608b"><div class="ttname"><a href="namespacetvm_1_1topi.html#a4435c7d12111817c485ebe9d2a6d608b">tvm::topi::fast_erf_float16</a></div><div class="ttdeci">Tensor fast_erf_float16(const Tensor &data, std::string name, std::string tag)</div><div class="ttdoc">Fast_erf_float expression from Eigen for float16. </div><div class="ttdef"><b>Definition:</b> elemwise.h:471</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a472208fa01448c0ed649bec218eed7f9"><div class="ttname"><a href="namespacetvm_1_1topi.html#a472208fa01448c0ed649bec218eed7f9">tvm::topi::acos</a></div><div class="ttdeci">Tensor acos(const Tensor &x, std::string name="T_" "acos", std::string tag=kElementWise)</div><div class="ttdef"><b>Definition:</b> elemwise.h:67</div></div>
diff --git a/docs/reference/api/doxygen/group__norm_8h_source.html b/docs/reference/api/doxygen/group__norm_8h_source.html
index 1149e211bc..de035821fd 100644
--- a/docs/reference/api/doxygen/group__norm_8h_source.html
+++ b/docs/reference/api/doxygen/group__norm_8h_source.html
@@ -79,7 +79,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html">tvm::runtime::Array</a></div><div class="ttdoc">Array, container representing a contiguous sequence of ObjectRefs. </div><div class="ttdef"><b>Definition:</b> array.h:289</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a5a5d59f85892da838847bced0990cb38"><div class="ttname"><a href="namespacetvm_1_1topi.html#a5a5d59f85892da838847bced0990cb38">tvm::topi::multiply</a></div><div class="ttdeci">tvm::PrimExpr multiply(const tvm::PrimExpr &a, const tvm::PrimExpr &b)</div><div class="ttdef"><b>Definition:</b> broadcast.h:225</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_aa7d7e1d5966e6de6b29c902458a421e8"><div class="ttname"><a href="namespacetvm_1_1topi.html#aa7d7e1d5966e6de6b29c902458a421e8">tvm::topi::add</a></div><div class="ttdeci">tvm::PrimExpr add(const tvm::PrimExpr &a, const tvm::PrimExpr &b)</div><div class="ttdef"><b>Definition:</b> broadcast.h:197</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a7b1acf424786ee187f0f19a725b85d8c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a7b1acf424786ee187f0f19a725b85d8c">tvm::topi::kCommReduce</a></div><div class="ttdeci">constexpr auto kCommReduce</div><div class="ttdef"><b>Definition:</b> tags.h:34</div></div>
<div class="ttc" id="namespacetvm_1_1te_html_ae0c71f84710b436cbe0b32289d0838f4"><div class="ttname"><a href="namespacetvm_1_1te.html#ae0c71f84710b436cbe0b32289d0838f4">tvm::te::var</a></div><div class="ttdeci">Var var(std::string name_hint, DataType t=DataType::Int(32))</div><div class="ttdoc">Construct a new Var expression. </div></div>
<div class="ttc" id="namespacetvm_html_af347f10e3572adb2d74ba4a53777db2b"><div class="ttname"><a href="namespacetvm.html#af347f10e3572adb2d74ba4a53777db2b">tvm::floordiv</a></div><div class="ttdeci">PrimExpr floordiv(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">compute floor(a / b) </div></div>
diff --git a/docs/reference/api/doxygen/index__map_8h_source.html b/docs/reference/api/doxygen/index__map_8h_source.html
index 8b9fe3fb0a..83a9efe600 100644
--- a/docs/reference/api/doxygen/index__map_8h_source.html
+++ b/docs/reference/api/doxygen/index__map_8h_source.html
@@ -89,7 +89,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1tir_1_1IndexMap_html"><div class="ttname"><a href="classtvm_1_1tir_1_1IndexMap.html">tvm::tir::IndexMap</a></div><div class="ttdef"><b>Definition:</b> index_map.h:177</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1String_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1String.html">tvm::runtime::String</a></div><div class="ttdoc">Reference to string objects. </div><div class="ttdef"><b>Definition:</b> string.h:98</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1TypedPackedFunc_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1TypedPackedFunc.html">tvm::runtime::TypedPackedFunc</a></div><div class="ttdoc">Please refer to TypedPackedFunc<R(Args..)>. </div><div class="ttdef"><b>Definition:</b> packed_func.h:60</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="object_8h_html_ac6e7295a4999e2c8e4a2c990beca887a"><div class="ttname"><a href="object_8h.html#ac6e7295a4999e2c8e4a2c990beca887a">TVM_DEFINE_OBJECT_REF_METHODS</a></div><div class="ttdeci">#define TVM_DEFINE_OBJECT_REF_METHODS(TypeName, ParentType, ObjectName)</div><div class="ttdef"><b>Definition:</b> object.h:713</div></div>
<div class="ttc" id="namespacetvm_1_1te_html_ae0c71f84710b436cbe0b32289d0838f4"><div class="ttname"><a href="namespacetvm_1_1te.html#ae0c71f84710b436cbe0b32289d0838f4">tvm::te::var</a></div><div class="ttdeci">Var var(std::string name_hint, DataType t=DataType::Int(32))</div><div class="ttdoc">Construct a new Var expression. </div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ObjectRef_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></div><div class="ttdoc">Base class of all object reference. </div><div class="ttdef"><b>Definition:</b> object.h:511</div></div>
diff --git a/docs/reference/api/doxygen/memory__manager_8h_source.html b/docs/reference/api/doxygen/memory__manager_8h_source.html
index 366b06bf1c..3837fe8d79 100644
--- a/docs/reference/api/doxygen/memory__manager_8h_source.html
+++ b/docs/reference/api/doxygen/memory__manager_8h_source.html
@@ -81,7 +81,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1runtime_1_1vm_1_1StorageObj_html_ae787a0a25d67ff1952ac5e9eeb0d1a61"><div class="ttname"><a href="classtvm_1_1runtime_1_1vm_1_1StorageObj.html#ae787a0a25d67ff1952ac5e9eeb0d1a61">tvm::runtime::vm::StorageObj::buffer</a></div><div class="ttdeci">Buffer buffer</div><div class="ttdoc">The index into the VM function table. </div><div class="ttdef"><b>Definition:</b> memory_manager.h:117</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1vm_1_1MemoryManager_html_aeaa1e35590fd9bcbceea389d3dc1c3c3"><div class="ttname"><a href="classtvm_1_1runtime_1_1vm_1_1MemoryManager.html#aeaa1e35590fd9bcbceea389d3dc1c3c3">tvm::runtime::vm::MemoryManager::GetAllocator</a></div><div class="ttdeci">static Allocator * GetAllocator(Device dev)</div><div class="ttdoc">Get an allocator given the context. </div></div>
<div class="ttc" id="structtvm_1_1runtime_1_1vm_1_1Buffer_html_a99624b7e535aba1bf5131a5afc08c1c4"><div class="ttname"><a href="structtvm_1_1runtime_1_1vm_1_1Buffer.html#a99624b7e535aba1bf5131a5afc08c1c4">tvm::runtime::vm::Buffer::data</a></div><div class="ttdeci">void * data</div><div class="ttdoc">The pointer to the allocated block of memory. </div><div class="ttdef"><b>Definition:</b> memory_manager.h:43</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="namespacetvm_html_a7c2095aed90b2129ba631b90103313a2"><div class="ttname"><a href="namespacetvm.html#a7c2095aed90b2129ba631b90103313a2">tvm::Device</a></div><div class="ttdeci">DLDevice Device</div><div class="ttdef"><b>Definition:</b> ndarray.h:43</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ObjectRef_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></div><div class="ttdoc">Base class of all object reference. </div><div class="ttdef"><b>Definition:</b> object.h:511</div></div>
<div class="ttc" id="object_8h_html"><div class="ttname"><a href="object_8h.html">object.h</a></div><div class="ttdoc">A managed object in the TVM runtime. </div></div>
diff --git a/docs/reference/api/doxygen/namespacemembers_func_l.html b/docs/reference/api/doxygen/namespacemembers_func_l.html
index 1716ada74f..9031420dfb 100644
--- a/docs/reference/api/doxygen/namespacemembers_func_l.html
+++ b/docs/reference/api/doxygen/namespacemembers_func_l.html
@@ -74,7 +74,7 @@ $(function() {
: <a class="el" href="namespacetvm_1_1topi_1_1nn.html#accc17ec14bd2ccc2c0e8055165ba69bd">tvm::topi::nn</a>
</li>
<li>layout_transform()
-: <a class="el" href="namespacetvm_1_1topi.html#a8a41a08eee70607889b738946ed97873">tvm::topi</a>
+: <a class="el" href="namespacetvm_1_1topi.html#a9a4108863737f5022a539a45d8ad59f0">tvm::topi</a>
</li>
<li>LazyGradientInit()
: <a class="el" href="namespacetvm_1_1relay_1_1transform.html#ab6e60251741cca1e811564b2f03a4a5d">tvm::relay::transform</a>
diff --git a/docs/reference/api/doxygen/namespacemembers_l.html b/docs/reference/api/doxygen/namespacemembers_l.html
index 25f97e8385..e06d15a785 100644
--- a/docs/reference/api/doxygen/namespacemembers_l.html
+++ b/docs/reference/api/doxygen/namespacemembers_l.html
@@ -77,7 +77,7 @@ $(function() {
: <a class="el" href="namespacetvm_1_1tir_1_1attr.html#a9cf212844cfa05381db84b0c470318e3">tvm::tir::attr</a>
</li>
<li>layout_transform()
-: <a class="el" href="namespacetvm_1_1topi.html#a8a41a08eee70607889b738946ed97873">tvm::topi</a>
+: <a class="el" href="namespacetvm_1_1topi.html#a9a4108863737f5022a539a45d8ad59f0">tvm::topi</a>
</li>
<li>layout_transforms
: <a class="el" href="namespacetvm_1_1tir_1_1attr.html#ace6dc58c76a20757b54c5afd76bdbf53">tvm::tir::attr</a>
diff --git a/docs/reference/api/doxygen/namespacetvm_1_1topi.html b/docs/reference/api/doxygen/namespacetvm_1_1topi.html
index 579119b112..d9a9408e1e 100644
--- a/docs/reference/api/doxygen/namespacetvm_1_1topi.html
+++ b/docs/reference/api/doxygen/namespacetvm_1_1topi.html
@@ -692,9 +692,9 @@ Functions</h2></td></tr>
<tr class="memitem:a453f9f5a35a9852a086ce0228ee22164"><td class="memItemLeft" align="right" valign="top"><a class="el" href="classtvm_1_1runtime_1_1Array.html">Array</a>< <a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> > </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacetvm_1_1topi.html#a453f9f5a35a9852a086ce0228ee22164">meshgrid</a> (const <a class="el" href="classtvm_1_1runtime_1_1Array.html">Array</a>< <a class="el" href="classtvm_1 [...]
<tr class="memdesc:a453f9f5a35a9852a086ce0228ee22164"><td class="mdescLeft"> </td><td class="mdescRight">Produce grids by expanding input over dimensions defined by other inputs. <a href="#a453f9f5a35a9852a086ce0228ee22164">More...</a><br /></td></tr>
<tr class="separator:a453f9f5a35a9852a086ce0228ee22164"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a8a41a08eee70607889b738946ed97873"><td class="memItemLeft" align="right" valign="top"><a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacetvm_1_1topi.html#a8a41a08eee70607889b738946ed97873">layout_transform</a> (const <a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> &src, const std::string &src_layout, const std::string &dst_layout, const std::string name= [...]
-<tr class="memdesc:a8a41a08eee70607889b738946ed97873"><td class="mdescLeft"> </td><td class="mdescRight">Transform the layout according to <code>src_layout</code> and <code>dst_layout</code>. <a href="#a8a41a08eee70607889b738946ed97873">More...</a><br /></td></tr>
-<tr class="separator:a8a41a08eee70607889b738946ed97873"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a9a4108863737f5022a539a45d8ad59f0"><td class="memItemLeft" align="right" valign="top"><a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacetvm_1_1topi.html#a9a4108863737f5022a539a45d8ad59f0">layout_transform</a> (const <a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> &src, const std::string &src_layout, const std::string &dst_layout, const std::string sched [...]
+<tr class="memdesc:a9a4108863737f5022a539a45d8ad59f0"><td class="mdescLeft"> </td><td class="mdescRight">Transform the layout according to <code>src_layout</code> and <code>dst_layout</code>. <a href="#a9a4108863737f5022a539a45d8ad59f0">More...</a><br /></td></tr>
+<tr class="separator:a9a4108863737f5022a539a45d8ad59f0"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:a6fec3a88d419cd018cf1041cf3d99204"><td class="memItemLeft" align="right" valign="top">void </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacetvm_1_1topi.html#a6fec3a88d419cd018cf1041cf3d99204">parse_auto_scheduler_layout</a> (const <a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> &layout, <a class="el" href="classtvm_1_1runtime_1_1Array.html">Array</a>< <a class="el" href="classtvm_1_1PrimExpr.html">PrimExpr</a> &g [...]
<tr class="memdesc:a6fec3a88d419cd018cf1041cf3d99204"><td class="mdescLeft"> </td><td class="mdescRight">Utility function for auto_scheduler_layout_transform. <a href="#a6fec3a88d419cd018cf1041cf3d99204">More...</a><br /></td></tr>
<tr class="separator:a6fec3a88d419cd018cf1041cf3d99204"><td class="memSeparator" colspan="2"> </td></tr>
@@ -5988,8 +5988,8 @@ Variables</h2></td></tr>
</div>
</div>
-<a id="a8a41a08eee70607889b738946ed97873"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a8a41a08eee70607889b738946ed97873">◆ </a></span>layout_transform()</h2>
+<a id="a9a4108863737f5022a539a45d8ad59f0"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a9a4108863737f5022a539a45d8ad59f0">◆ </a></span>layout_transform()</h2>
<div class="memitem">
<div class="memproto">
@@ -6015,6 +6015,12 @@ Variables</h2></td></tr>
<td class="paramtype">const std::string & </td>
<td class="paramname"><em>dst_layout</em>, </td>
</tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">const std::string </td>
+ <td class="paramname"><em>schedule_rule</em> = <code>"None"</code>, </td>
+ </tr>
<tr>
<td class="paramkey"></td>
<td></td>
@@ -6048,6 +6054,7 @@ Variables</h2></td></tr>
<tr><td class="paramname">dst_layout</td><td>the destination layout. </td></tr>
<tr><td class="paramname">name</td><td>output tensor name. </td></tr>
<tr><td class="paramname">tag</td><td>output tensor tag. </td></tr>
+ <tr><td class="paramname">schedule_rule</td><td>name of specialized schedule rule to use. </td></tr>
</table>
</dd>
</dl>
diff --git a/docs/reference/api/doxygen/ndarray_8h_source.html b/docs/reference/api/doxygen/ndarray_8h_source.html
index 104ca6a75f..d48e71b8d6 100644
--- a/docs/reference/api/doxygen/ndarray_8h_source.html
+++ b/docs/reference/api/doxygen/ndarray_8h_source.html
@@ -111,7 +111,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1runtime_1_1ObjectRef_html_ac261cdb80487fb29ac42b28678f8cbef"><div class="ttname"><a href="classtvm_1_1runtime_1_1ObjectRef.html#ac261cdb80487fb29ac42b28678f8cbef">tvm::runtime::ObjectRef::data_</a></div><div class="ttdeci">ObjectPtr< Object > data_</div><div class="ttdoc">Internal pointer that backs the reference. </div><div class="ttdef"><b>Definition:</b> object.h:574</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1NDArray_1_1ContainerBase_html_a1063a9d01075d5b7b0e8fa31d4d72e0b"><div class="ttname"><a href="classtvm_1_1runtime_1_1NDArray_1_1ContainerBase.html#a1063a9d01075d5b7b0e8fa31d4d72e0b">tvm::runtime::NDArray::ContainerBase::dl_tensor</a></div><div class="ttdeci">DLTensor dl_tensor</div><div class="ttdoc">The corresponding dl_tensor field. </div><div class="ttdef"><b>Definition:</b> ndarray.h:264</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1NDArray_1_1Container_html_a56109cfc826b26172f084c3790144351"><div class="ttname"><a href="classtvm_1_1runtime_1_1NDArray_1_1Container.html#a56109cfc826b26172f084c3790144351">tvm::runtime::NDArray::Container::SetDeleter</a></div><div class="ttdeci">void SetDeleter(FDeleter deleter)</div><div class="ttdoc">Set the deleter field. </div><div class="ttdef"><b>Definition:</b> ndarray.h:315</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="c__runtime__api_8h_html_ace8007daffd9f2c6d954c24d870bfcc4"><div class="ttname"><a href="c__runtime__api_8h.html#ace8007daffd9f2c6d954c24d870bfcc4">tvm_index_t</a></div><div class="ttdeci">int64_t tvm_index_t</div><div class="ttdoc">type of array index. </div><div class="ttdef"><b>Definition:</b> c_runtime_api.h:88</div></div>
<div class="ttc" id="namespacetvm_html_a7c2095aed90b2129ba631b90103313a2"><div class="ttname"><a href="namespacetvm.html#a7c2095aed90b2129ba631b90103313a2">tvm::Device</a></div><div class="ttdeci">DLDevice Device</div><div class="ttdef"><b>Definition:</b> ndarray.h:43</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ObjectRef_html_aadbc0886ffa80162ff31eefd0431ba09"><div class="ttname"><a href="classtvm_1_1runtime_1_1ObjectRef.html#aadbc0886ffa80162ff31eefd0431ba09">tvm::runtime::ObjectRef::get</a></div><div class="ttdeci">const Object * get() const</div><div class="ttdef"><b>Definition:</b> object.h:546</div></div>
diff --git a/docs/reference/api/doxygen/operation_8h_source.html b/docs/reference/api/doxygen/operation_8h_source.html
index f518a95ce3..5287e0ea97 100644
--- a/docs/reference/api/doxygen/operation_8h_source.html
+++ b/docs/reference/api/doxygen/operation_8h_source.html
@@ -122,7 +122,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1te_1_1ExternOp_html"><div class="ttname"><a href="classtvm_1_1te_1_1ExternOp.html">tvm::te::ExternOp</a></div><div class="ttdoc">Managed reference to ExternOpNode. </div><div class="ttdef"><b>Definition:</b> operation.h:460</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1String_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1String.html">tvm::runtime::String</a></div><div class="ttdoc">Reference to string objects. </div><div class="ttdef"><b>Definition:</b> string.h:98</div></div>
<div class="ttc" id="classtvm_1_1te_1_1TensorComputeOpNode_html_a81533c6957c82df59ef8d810ace1bed2"><div class="ttname"><a href="classtvm_1_1te_1_1TensorComputeOpNode.html#a81533c6957c82df59ef8d810ace1bed2">tvm::te::TensorComputeOpNode::inputs</a></div><div class="ttdeci">Array< Tensor > inputs</div><div class="ttdoc">input tensors of intrin </div><div class="ttdef"><b>Definition:</b> operation.h:281</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="object_8h_html_ac6e7295a4999e2c8e4a2c990beca887a"><div class="ttname"><a href="object_8h.html#ac6e7295a4999e2c8e4a2c990beca887a">TVM_DEFINE_OBJECT_REF_METHODS</a></div><div class="ttdeci">#define TVM_DEFINE_OBJECT_REF_METHODS(TypeName, ParentType, ObjectName)</div><div class="ttdef"><b>Definition:</b> object.h:713</div></div>
<div class="ttc" id="classtvm_1_1te_1_1ScanOp_html"><div class="ttname"><a href="classtvm_1_1te_1_1ScanOp.html">tvm::te::ScanOp</a></div><div class="ttdoc">Managed reference to ScanOpNode. </div><div class="ttdef"><b>Definition:</b> operation.h:399</div></div>
<div class="ttc" id="classtvm_1_1te_1_1ScanOpNode_html_a1b681295f74cb94732ef167a15a8488f"><div class="ttname"><a href="classtvm_1_1te_1_1ScanOpNode.html#a1b681295f74cb94732ef167a15a8488f">tvm::te::ScanOpNode::ScanOpNode</a></div><div class="ttdeci">ScanOpNode()</div><div class="ttdoc">constructor </div><div class="ttdef"><b>Definition:</b> operation.h:360</div></div>
diff --git a/docs/reference/api/doxygen/profiling_8h_source.html b/docs/reference/api/doxygen/profiling_8h_source.html
index a2b156db22..1309c3e6af 100644
--- a/docs/reference/api/doxygen/profiling_8h_source.html
+++ b/docs/reference/api/doxygen/profiling_8h_source.html
@@ -102,7 +102,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1runtime_1_1String_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1String.html">tvm::runtime::String</a></div><div class="ttdoc">Reference to string objects. </div><div class="ttdef"><b>Definition:</b> string.h:98</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1profiling_1_1ReportNode_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1profiling_1_1ReportNode.html">tvm::runtime::profiling::ReportNode</a></div><div class="ttdoc">Data collected from a profiling run. Includes per-call metrics and per-device metrics. </div><div class="ttdef"><b>Definition:</b> profiling.h:178</div></div>
<div class="ttc" id="device__api_8h_html"><div class="ttname"><a href="device__api_8h.html">device_api.h</a></div><div class="ttdoc">Abstract device memory management API. </div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="namespacetvm_html_a7c2095aed90b2129ba631b90103313a2"><div class="ttname"><a href="namespacetvm.html#a7c2095aed90b2129ba631b90103313a2">tvm::Device</a></div><div class="ttdeci">DLDevice Device</div><div class="ttdef"><b>Definition:</b> ndarray.h:43</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1profiling_1_1Report_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1profiling_1_1Report.html">tvm::runtime::profiling::Report</a></div><div class="ttdef"><b>Definition:</b> profiling.h:259</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1profiling_1_1ReportNode_html_a03216cd54383d55f809d2c1e6b6ac726"><div class="ttname"><a href="classtvm_1_1runtime_1_1profiling_1_1ReportNode.html#a03216cd54383d55f809d2c1e6b6ac726">tvm::runtime::profiling::ReportNode::configuration</a></div><div class="ttdeci">Map< String, ObjectRef > configuration</div><div class="ttdef"><b>Definition:</b> profiling.h:200</div></div>
diff --git a/docs/reference/api/doxygen/ravel__unravel_8h_source.html b/docs/reference/api/doxygen/ravel__unravel_8h_source.html
index fb4b8a999c..13fc53eeda 100644
--- a/docs/reference/api/doxygen/ravel__unravel_8h_source.html
+++ b/docs/reference/api/doxygen/ravel__unravel_8h_source.html
@@ -70,7 +70,7 @@ $(function() {
<div class="ttc" id="namespacetvm_html"><div class="ttname"><a href="namespacetvm.html">tvm</a></div><div class="ttdoc">runtime implementation for LibTorch/TorchScript. </div><div class="ttdef"><b>Definition:</b> analyzer.h:36</div></div>
<div class="ttc" id="namespacetvm_1_1te_html"><div class="ttname"><a href="namespacetvm_1_1te.html">tvm::te</a></div><div class="ttdoc">Tensor expression language DSL. </div><div class="ttdef"><b>Definition:</b> extracted_task.h:33</div></div>
<div class="ttc" id="namespacetvm_html_a8f30aa0685ca52f846843e76a1ad1dc7"><div class="ttname"><a href="namespacetvm.html#a8f30aa0685ca52f846843e76a1ad1dc7">tvm::indexdiv</a></div><div class="ttdeci">PrimExpr indexdiv(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">compute floor(a / b) where a and b are non-negative. </div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="operation_8h_html"><div class="ttname"><a href="operation_8h.html">operation.h</a></div><div class="ttdoc">Operation node can generate one or multiple Tensors. </div></div>
</div><!-- fragment --></div><!-- contents -->
<!-- start footer part -->
diff --git a/docs/reference/api/doxygen/search/all_d.js b/docs/reference/api/doxygen/search/all_d.js
index f0d5dc6050..ec6114fd31 100644
--- a/docs/reference/api/doxygen/search/all_d.js
+++ b/docs/reference/api/doxygen/search/all_d.js
@@ -22,7 +22,7 @@ var searchData=
['layout_5ffree_5fbuffers',['layout_free_buffers',['../namespacetvm_1_1tir_1_1attr.html#a9cf212844cfa05381db84b0c470318e3',1,'tvm::tir::attr']]],
['layout_5ffree_5fplaceholders_5fkey',['layout_free_placeholders_key',['../classtvm_1_1auto__scheduler_1_1ComputeDAG.html#a173f178286d543c71feb0f8a06d0cb7b',1,'tvm::auto_scheduler::ComputeDAG']]],
['layout_5frewrite_5foption',['layout_rewrite_option',['../classtvm_1_1auto__scheduler_1_1SearchTaskNode.html#ad770dabf0716145fe86537b9107856aa',1,'tvm::auto_scheduler::SearchTaskNode']]],
- ['layout_5ftransform',['layout_transform',['../namespacetvm_1_1topi.html#a8a41a08eee70607889b738946ed97873',1,'tvm::topi']]],
+ ['layout_5ftransform',['layout_transform',['../namespacetvm_1_1topi.html#a9a4108863737f5022a539a45d8ad59f0',1,'tvm::topi']]],
['layout_5ftransforms',['layout_transforms',['../classtvm_1_1te_1_1StageNode.html#a99d637a0da3b9f5d688f62410c884bea',1,'tvm::te::StageNode::layout_transforms()'],['../namespacetvm_1_1tir_1_1attr.html#ace6dc58c76a20757b54c5afd76bdbf53',1,'tvm::tir::attr::layout_transforms()']]],
['layoutaxis',['LayoutAxis',['../classtvm_1_1tir_1_1LayoutAxis.html',1,'tvm::tir']]],
['layoutnode',['LayoutNode',['../classtvm_1_1tir_1_1LayoutNode.html',1,'tvm::tir']]],
diff --git a/docs/reference/api/doxygen/search/functions_c.js b/docs/reference/api/doxygen/search/functions_c.js
index 55b20b7b79..8a6d50866f 100644
--- a/docs/reference/api/doxygen/search/functions_c.js
+++ b/docs/reference/api/doxygen/search/functions_c.js
@@ -9,7 +9,7 @@ var searchData=
['launchthread',['LaunchThread',['../namespacetvm_1_1script_1_1ir__builder_1_1tir.html#a1b6a75a0308366b04fee2fe37a22a297',1,'tvm::script::ir_builder::tir::LaunchThread(Var var, PrimExpr extent)'],['../namespacetvm_1_1script_1_1ir__builder_1_1tir.html#a6816c03cbd3c8baf2e27b94c381ca5fe',1,'tvm::script::ir_builder::tir::LaunchThread(String thread_tag, PrimExpr extent)']]],
['layer_5fnorm',['layer_norm',['../namespacetvm_1_1topi_1_1nn.html#accc17ec14bd2ccc2c0e8055165ba69bd',1,'tvm::topi::nn']]],
['layout',['Layout',['../classtvm_1_1tir_1_1Layout.html#ace4d6c608f7148bb2c6aefe33b4e4ab5',1,'tvm::tir::Layout::Layout(const Array< tir::IterVar > &axes)'],['../classtvm_1_1tir_1_1Layout.html#ac899b7c62ee68d8b882a119ba6854f88',1,'tvm::tir::Layout::Layout(const tvm::String &name)'],['../classtvm_1_1tir_1_1Layout.html#a11a8465c3cc562e408a48a7f0fd324fd',1,'tvm::tir::Layout::Layout(const char *name)'],['../classtvm_1_1tir_1_1Layout.html#acd20f31024c5a15e8109c64109582028',1,'t [...]
- ['layout_5ftransform',['layout_transform',['../namespacetvm_1_1topi.html#a8a41a08eee70607889b738946ed97873',1,'tvm::topi']]],
+ ['layout_5ftransform',['layout_transform',['../namespacetvm_1_1topi.html#a9a4108863737f5022a539a45d8ad59f0',1,'tvm::topi']]],
['lazygradientinit',['LazyGradientInit',['../namespacetvm_1_1relay_1_1transform.html#ab6e60251741cca1e811564b2f03a4a5d',1,'tvm::relay::transform']]],
['ldexp',['ldexp',['../namespacetvm.html#ac5347541411e75f59758a29596565f63',1,'tvm']]],
['le',['LE',['../classtvm_1_1tir_1_1LE.html#aa55713e8c1af53b5442602821c1da5fa',1,'tvm::tir::LE']]],
diff --git a/docs/reference/api/doxygen/shape__tuple_8h_source.html b/docs/reference/api/doxygen/shape__tuple_8h_source.html
index 1d67dd0cf1..cf4dafef90 100644
--- a/docs/reference/api/doxygen/shape__tuple_8h_source.html
+++ b/docs/reference/api/doxygen/shape__tuple_8h_source.html
@@ -80,7 +80,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1runtime_1_1ShapeTuple_html_a6836ad757cbcac207ddf025a74ac2ef2"><div class="ttname"><a href="classtvm_1_1runtime_1_1ShapeTuple.html#a6836ad757cbcac207ddf025a74ac2ef2">tvm::runtime::ShapeTuple::ShapeTuple</a></div><div class="ttdeci">ShapeTuple(IterType begin, IterType end)</div><div class="ttdoc">Constructor from iterator. </div><div class="ttdef"><b>Definition:</b> shape_tuple.h:98</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ShapeTuple_html_a93afa6c6de6d38665b62a007892a8632"><div class="ttname"><a href="classtvm_1_1runtime_1_1ShapeTuple.html#a93afa6c6de6d38665b62a007892a8632">tvm::runtime::ShapeTuple::begin</a></div><div class="ttdeci">const index_type * begin() const</div><div class="ttdef"><b>Definition:</b> shape_tuple.h:158</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ShapeTuple_html_a7e4f6df331fcf589c3f5544c88931561"><div class="ttname"><a href="classtvm_1_1runtime_1_1ShapeTuple.html#a7e4f6df331fcf589c3f5544c88931561">tvm::runtime::ShapeTuple::empty</a></div><div class="ttdeci">bool empty() const</div><div class="ttdef"><b>Definition:</b> shape_tuple.h:149</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ShapeTupleObj_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1ShapeTupleObj.html">tvm::runtime::ShapeTupleObj</a></div><div class="ttdoc">An object representing a shape tuple. </div><div class="ttdef"><b>Definition:</b> shape_tuple.h:36</div></div>
<div class="ttc" id="structtvm_1_1runtime_1_1TypeIndex_html_aed93c7318efc8052201d4c404b21a40dae2bdfcaf2a93d7e1dfef9012bd0cae77"><div class="ttname"><a href="structtvm_1_1runtime_1_1TypeIndex.html#aed93c7318efc8052201d4c404b21a40dae2bdfcaf2a93d7e1dfef9012bd0cae77">tvm::runtime::TypeIndex::kRuntimeShapeTuple</a></div><div class="ttdoc">runtime::ShapeTuple. </div><div class="ttdef"><b>Definition:</b> object.h:72</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ObjectRef_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></div><div class="ttdoc">Base class of all object reference. </div><div class="ttdef"><b>Definition:</b> object.h:511</div></div>
diff --git a/docs/reference/api/doxygen/tensor_8h_source.html b/docs/reference/api/doxygen/tensor_8h_source.html
index f853356a75..c3d55d1fac 100644
--- a/docs/reference/api/doxygen/tensor_8h_source.html
+++ b/docs/reference/api/doxygen/tensor_8h_source.html
@@ -95,7 +95,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1runtime_1_1String_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1String.html">tvm::runtime::String</a></div><div class="ttdoc">Reference to string objects. </div><div class="ttdef"><b>Definition:</b> string.h:98</div></div>
<div class="ttc" id="classtvm_1_1te_1_1TensorNode_html_a153569448cb1bf9d2924d35639c3b8b8"><div class="ttname"><a href="classtvm_1_1te_1_1TensorNode.html#a153569448cb1bf9d2924d35639c3b8b8">tvm::te::TensorNode::TensorNode</a></div><div class="ttdeci">TensorNode()</div><div class="ttdoc">constructor </div><div class="ttdef"><b>Definition:</b> tensor.h:79</div></div>
<div class="ttc" id="classtvm_1_1te_1_1Tensor_1_1Slice_html_a1a00e3ec5f80973c337b0e7ab9c0974d"><div class="ttname"><a href="classtvm_1_1te_1_1Tensor_1_1Slice.html#a1a00e3ec5f80973c337b0e7ab9c0974d">tvm::te::Tensor::Slice::operator[]</a></div><div class="ttdeci">Slice operator[](PrimExpr i)</div><div class="ttdoc">get i-th slice from the current slice. </div><div class="ttdef"><b>Definition:</b> tensor.h:187</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="object_8h_html_ac6e7295a4999e2c8e4a2c990beca887a"><div class="ttname"><a href="object_8h.html#ac6e7295a4999e2c8e4a2c990beca887a">TVM_DEFINE_OBJECT_REF_METHODS</a></div><div class="ttdeci">#define TVM_DEFINE_OBJECT_REF_METHODS(TypeName, ParentType, ObjectName)</div><div class="ttdef"><b>Definition:</b> object.h:713</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ObjectRef_html_aadbc0886ffa80162ff31eefd0431ba09"><div class="ttname"><a href="classtvm_1_1runtime_1_1ObjectRef.html#aadbc0886ffa80162ff31eefd0431ba09">tvm::runtime::ObjectRef::get</a></div><div class="ttdeci">const Object * get() const</div><div class="ttdef"><b>Definition:</b> object.h:546</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1ObjectRef_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></div><div class="ttdoc">Base class of all object reference. </div><div class="ttdef"><b>Definition:</b> object.h:511</div></div>
diff --git a/docs/reference/api/doxygen/tensor__type_8h_source.html b/docs/reference/api/doxygen/tensor__type_8h_source.html
index a230030a6b..b62bc360ef 100644
--- a/docs/reference/api/doxygen/tensor__type_8h_source.html
+++ b/docs/reference/api/doxygen/tensor__type_8h_source.html
@@ -88,7 +88,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1TensorTypeNode_html_a84759da6fdc0984edcc65a8dda484505"><div class="ttname"><a href="classtvm_1_1TensorTypeNode.html#a84759da6fdc0984edcc65a8dda484505">tvm::TensorTypeNode::SEqualReduce</a></div><div class="ttdeci">bool SEqualReduce(const TensorTypeNode *other, SEqualReducer equal) const</div><div class="ttdef"><b>Definition:</b> tensor_type.h:77</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html">tvm::runtime::Array</a></div><div class="ttdoc">Array, container representing a contiguous sequence of ObjectRefs. </div><div class="ttdef"><b>Definition:</b> array.h:289</div></div>
<div class="ttc" id="classtvm_1_1TypeNode_html"><div class="ttname"><a href="classtvm_1_1TypeNode.html">tvm::TypeNode</a></div><div class="ttdoc">Type is the base type of all types. </div><div class="ttdef"><b>Definition:</b> type.h:74</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="classtvm_1_1BaseTensorTypeNode_html_afc675db92c700532b65fe819b1a37eb5"><div class="ttname"><a href="classtvm_1_1BaseTensorTypeNode.html#afc675db92c700532b65fe819b1a37eb5">tvm::BaseTensorTypeNode::TVM_DECLARE_BASE_OBJECT_INFO</a></div><div class="ttdeci">TVM_DECLARE_BASE_OBJECT_INFO(BaseTensorTypeNode, TypeNode)</div></div>
<div class="ttc" id="object_8h_html_ac6e7295a4999e2c8e4a2c990beca887a"><div class="ttname"><a href="object_8h.html#ac6e7295a4999e2c8e4a2c990beca887a">TVM_DEFINE_OBJECT_REF_METHODS</a></div><div class="ttdeci">#define TVM_DEFINE_OBJECT_REF_METHODS(TypeName, ParentType, ObjectName)</div><div class="ttdef"><b>Definition:</b> object.h:713</div></div>
<div class="ttc" id="object_8h_html_a3aea9b3f65aeb9150c0fa7800e5573c6"><div class="ttname"><a href="object_8h.html#a3aea9b3f65aeb9150c0fa7800e5573c6">TVM_DECLARE_FINAL_OBJECT_INFO</a></div><div class="ttdeci">#define TVM_DECLARE_FINAL_OBJECT_INFO(TypeName, ParentType)</div><div class="ttdoc">helper macro to declare type information in a final class. </div><div class="ttdef"><b>Definition:</b> object.h:671</div></div>
diff --git a/docs/reference/api/doxygen/tir_2ir_8h_source.html b/docs/reference/api/doxygen/tir_2ir_8h_source.html
index 2c09a3fde9..5f32a29ccf 100644
--- a/docs/reference/api/doxygen/tir_2ir_8h_source.html
+++ b/docs/reference/api/doxygen/tir_2ir_8h_source.html
@@ -117,7 +117,7 @@ $(function() {
<div class="ttc" id="namespacetvm_1_1script_1_1ir__builder_1_1tir_html_a6df09cb19f1e26c3fa0e1e0251c3c0bb"><div class="ttname"><a href="namespacetvm_1_1script_1_1ir__builder_1_1tir.html#a6df09cb19f1e26c3fa0e1e0251c3c0bb">tvm::script::ir_builder::tir::While</a></div><div class="ttdeci">WhileFrame While(PrimExpr condition)</div><div class="ttdoc">Create a while loop. </div></div>
<div class="ttc" id="namespacetvm_1_1script_1_1ir__builder_1_1tir_html_a454a28cc9ed56389b7b09b5b45a3097e"><div class="ttname"><a href="namespacetvm_1_1script_1_1ir__builder_1_1tir.html#a454a28cc9ed56389b7b09b5b45a3097e">tvm::script::ir_builder::tir::Where</a></div><div class="ttdeci">void Where(PrimExpr predicate)</div><div class="ttdoc">The block predicate statement. </div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1String_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1String.html">tvm::runtime::String</a></div><div class="ttdoc">Reference to string objects. </div><div class="ttdef"><b>Definition:</b> string.h:98</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="namespacetvm_1_1script_1_1ir__builder_1_1tir_html_af443e6c0816e4e4ff97881df7a11a6db"><div class="ttname"><a href="namespacetvm_1_1script_1_1ir__builder_1_1tir.html#af443e6c0816e4e4ff97881df7a11a6db">tvm::script::ir_builder::tir::Parallel</a></div><div class="ttdeci">ForFrame Parallel(PrimExpr start, PrimExpr stop, Optional< Map< String, ObjectRef >> annotations=NullOpt)</div><div class="ttdoc">The parallel For statement. </div></div>
<div class="ttc" id="namespacetvm_1_1relay_html_a661d95f170bca230773914caeef3fe52"><div class="ttname"><a href="namespacetvm_1_1relay.html#a661d95f170bca230773914caeef3fe52">tvm::relay::Type</a></div><div class="ttdeci">tvm::Type Type</div><div class="ttdef"><b>Definition:</b> type.h:47</div></div>
<div class="ttc" id="namespacetvm_1_1te_html_ae0c71f84710b436cbe0b32289d0838f4"><div class="ttname"><a href="namespacetvm_1_1te.html#ae0c71f84710b436cbe0b32289d0838f4">tvm::te::var</a></div><div class="ttdeci">Var var(std::string name_hint, DataType t=DataType::Int(32))</div><div class="ttdoc">Construct a new Var expression. </div></div>
diff --git a/docs/reference/api/doxygen/topi_2nn_8h_source.html b/docs/reference/api/doxygen/topi_2nn_8h_source.html
index 196ba878b8..082efa1e94 100644
--- a/docs/reference/api/doxygen/topi_2nn_8h_source.html
+++ b/docs/reference/api/doxygen/topi_2nn_8h_source.html
@@ -103,7 +103,7 @@ $(function() {
<div class="ttc" id="namespacetvm_1_1topi_html_a0250c4095f19ae8a22ed85bc4ce5a40d"><div class="ttname"><a href="namespacetvm_1_1topi.html#a0250c4095f19ae8a22ed85bc4ce5a40d">tvm::topi::kElementWise</a></div><div class="ttdeci">constexpr auto kElementWise</div><div class="ttdef"><b>Definition:</b> tags.h:32</div></div>
<div class="ttc" id="namespacetvm_html_a8f30aa0685ca52f846843e76a1ad1dc7"><div class="ttname"><a href="namespacetvm.html#a8f30aa0685ca52f846843e76a1ad1dc7">tvm::indexdiv</a></div><div class="ttdeci">PrimExpr indexdiv(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">compute floor(a / b) where a and b are non-negative. </div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a786c950302117f8bdbbcab523f7b83bb"><div class="ttname"><a href="namespacetvm_1_1topi.html#a786c950302117f8bdbbcab523f7b83bb">tvm::topi::depthwise_conv2d_nchw</a></div><div class="ttdeci">tvm::te::Tensor depthwise_conv2d_nchw(const tvm::te::Tensor &I, const tvm::te::Tensor &W, int pad_h=0, int pad_w=0, int stride_h=1, int stride_w=1, std::string name="T_depthwise_conv2d_nchw", std::string tag=kDepthwiseConv2dNCHW)</div><div [...]
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="namespacetvm_1_1te_html_aae384e9b73c2271905486e4a74b69265"><div class="ttname"><a href="namespacetvm_1_1te.html#aae384e9b73c2271905486e4a74b69265">tvm::te::reduce_axis</a></div><div class="ttdeci">IterVar reduce_axis(Range dom, std::string name="rv")</div><div class="ttdoc">Create a new IterVar for reduction operations. </div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a3305d377f96cd20c23032eeada2756d5"><div class="ttname"><a href="namespacetvm_1_1topi.html#a3305d377f96cd20c23032eeada2756d5">tvm::topi::pad</a></div><div class="ttdeci">tvm::te::Tensor pad(const tvm::te::Tensor &t, const tvm::Array< tvm::PrimExpr > &pad_before, tvm::Array< tvm::PrimExpr > pad_after=tvm::Array< tvm::PrimExpr >(), PrimExpr pad_value=PrimExpr(), std::string name="T_pad", std::string tag=kElem [...]
<div class="ttc" id="namespacetvm_html_a27d5567b95675d383c4675fdcd85346c"><div class="ttname"><a href="namespacetvm.html#a27d5567b95675d383c4675fdcd85346c">tvm::logical_and</a></div><div class="ttdeci">PrimExpr logical_and(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">and </div></div>
diff --git a/docs/reference/api/doxygen/topi_2transform_8h.html b/docs/reference/api/doxygen/topi_2transform_8h.html
index 00199485f7..9fec76dd6e 100644
--- a/docs/reference/api/doxygen/topi_2transform_8h.html
+++ b/docs/reference/api/doxygen/topi_2transform_8h.html
@@ -202,9 +202,9 @@ Functions</h2></td></tr>
<tr class="memitem:a453f9f5a35a9852a086ce0228ee22164"><td class="memItemLeft" align="right" valign="top">Array< <a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> > </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacetvm_1_1topi.html#a453f9f5a35a9852a086ce0228ee22164">tvm::topi::meshgrid</a> (const Array< <a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> > &inputs, const std::string &indexing, std::string name="T [...]
<tr class="memdesc:a453f9f5a35a9852a086ce0228ee22164"><td class="mdescLeft"> </td><td class="mdescRight">Produce grids by expanding input over dimensions defined by other inputs. <a href="namespacetvm_1_1topi.html#a453f9f5a35a9852a086ce0228ee22164">More...</a><br /></td></tr>
<tr class="separator:a453f9f5a35a9852a086ce0228ee22164"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a8a41a08eee70607889b738946ed97873"><td class="memItemLeft" align="right" valign="top"><a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacetvm_1_1topi.html#a8a41a08eee70607889b738946ed97873">tvm::topi::layout_transform</a> (const <a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> &src, const std::string &src_layout, const std::string &dst_layout, const std::s [...]
-<tr class="memdesc:a8a41a08eee70607889b738946ed97873"><td class="mdescLeft"> </td><td class="mdescRight">Transform the layout according to <code>src_layout</code> and <code>dst_layout</code>. <a href="namespacetvm_1_1topi.html#a8a41a08eee70607889b738946ed97873">More...</a><br /></td></tr>
-<tr class="separator:a8a41a08eee70607889b738946ed97873"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a9a4108863737f5022a539a45d8ad59f0"><td class="memItemLeft" align="right" valign="top"><a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacetvm_1_1topi.html#a9a4108863737f5022a539a45d8ad59f0">tvm::topi::layout_transform</a> (const <a class="el" href="classtvm_1_1te_1_1Tensor.html">Tensor</a> &src, const std::string &src_layout, const std::string &dst_layout, const std::s [...]
+<tr class="memdesc:a9a4108863737f5022a539a45d8ad59f0"><td class="mdescLeft"> </td><td class="mdescRight">Transform the layout according to <code>src_layout</code> and <code>dst_layout</code>. <a href="namespacetvm_1_1topi.html#a9a4108863737f5022a539a45d8ad59f0">More...</a><br /></td></tr>
+<tr class="separator:a9a4108863737f5022a539a45d8ad59f0"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:a6fec3a88d419cd018cf1041cf3d99204"><td class="memItemLeft" align="right" valign="top">void </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacetvm_1_1topi.html#a6fec3a88d419cd018cf1041cf3d99204">tvm::topi::parse_auto_scheduler_layout</a> (const String &layout, Array< PrimExpr > *shape, std::vector< std::string > *axes)</td></tr>
<tr class="memdesc:a6fec3a88d419cd018cf1041cf3d99204"><td class="mdescLeft"> </td><td class="mdescRight">Utility function for auto_scheduler_layout_transform. <a href="namespacetvm_1_1topi.html#a6fec3a88d419cd018cf1041cf3d99204">More...</a><br /></td></tr>
<tr class="separator:a6fec3a88d419cd018cf1041cf3d99204"><td class="memSeparator" colspan="2"> </td></tr>
diff --git a/docs/reference/api/doxygen/topi_2transform_8h_source.html b/docs/reference/api/doxygen/topi_2transform_8h_source.html
index 1d2b39dd05..3f6c210ad8 100644
--- a/docs/reference/api/doxygen/topi_2transform_8h_source.html
+++ b/docs/reference/api/doxygen/topi_2transform_8h_source.html
@@ -66,27 +66,28 @@ $(function() {
<div class="title">transform.h</div> </div>
</div><!--header-->
<div class="contents">
-<a href="topi_2transform_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * or [...]
+<a href="topi_2transform_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * or [...]
<div class="ttc" id="classtvm_1_1tir_1_1Layout_html"><div class="ttname"><a href="classtvm_1_1tir_1_1Layout.html">tvm::tir::Layout</a></div><div class="ttdoc">Managed reference to LayoutNode. </div><div class="ttdef"><b>Definition:</b> data_layout.h:123</div></div>
<div class="ttc" id="namespacetvm_html_aac2abc149c1a47944c37b560181b15c0"><div class="ttname"><a href="namespacetvm.html#aac2abc149c1a47944c37b560181b15c0">tvm::min</a></div><div class="ttdeci">PrimExpr min(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">take minimum of two values </div></div>
<div class="ttc" id="classtvm_1_1tir_1_1Layout_html_a35389f5ca857483ea65756bb2ceaa944"><div class="ttname"><a href="classtvm_1_1tir_1_1Layout.html#a35389f5ca857483ea65756bb2ceaa944">tvm::tir::Layout::Equals</a></div><div class="ttdeci">bool Equals(const Layout &rhs) const</div><div class="ttdoc">Whether the two layouts are equal. </div><div class="ttdef"><b>Definition:</b> data_layout.h:278</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_acd9415b24624edac4edec4802d275062"><div class="ttname"><a href="namespacetvm_1_1topi.html#acd9415b24624edac4edec4802d275062">tvm::topi::strided_slice_with_axes</a></div><div class="ttdeci">Tensor strided_slice_with_axes(const Tensor &x, const Array< Integer > &begin, const Array< Integer > &end, const Array< Integer > &strides, const Array< Integer > &axes, std::string slice_mode="end", s [...]
-<div class="ttc" id="namespacetvm_1_1topi_html_a877e6fdffb6b6c051c29602ec6fe995c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a877e6fdffb6b6c051c29602ec6fe995c">tvm::topi::sparse_to_dense</a></div><div class="ttdeci">Tensor sparse_to_dense(const Tensor &sparse_indices, const Array< PrimExpr > &output_shape, const Tensor &sparse_values, const PrimExpr &default_value, const std::string name="T_sparse_to_dense", const std::string tag=kInjective)</d [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_a877e6fdffb6b6c051c29602ec6fe995c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a877e6fdffb6b6c051c29602ec6fe995c">tvm::topi::sparse_to_dense</a></div><div class="ttdeci">Tensor sparse_to_dense(const Tensor &sparse_indices, const Array< PrimExpr > &output_shape, const Tensor &sparse_values, const PrimExpr &default_value, const std::string name="T_sparse_to_dense", const std::string tag=kInjective)</d [...]
<div class="ttc" id="namespacetvm_html_ada5ad8338d3144221d8f16380e6c4855"><div class="ttname"><a href="namespacetvm.html#ada5ad8338d3144221d8f16380e6c4855">tvm::indexmod</a></div><div class="ttdeci">PrimExpr indexmod(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">compute the remainder floor(a / b) where a and b are non-negative. </div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_aff75b2e533b654909ca1bd9485ef4e6b"><div class="ttname"><a href="namespacetvm_1_1topi.html#aff75b2e533b654909ca1bd9485ef4e6b">tvm::topi::StridedSliceOutputShape</a></div><div class="ttdeci">Array< PrimExpr > StridedSliceOutputShape(const Array< PrimExpr > &ishape, const Array< Integer > &begin, const Array< Integer > &end, const Array< Integer > &strides, const Array< Integer > &axes, co [...]
<div class="ttc" id="namespacetvm_1_1tir_html_a1a071208bbbab6b220cf46f5cdccdd86"><div class="ttname"><a href="namespacetvm_1_1tir.html#a1a071208bbbab6b220cf46f5cdccdd86">tvm::tir::make_const</a></div><div class="ttdeci">PrimExpr make_const(DataType t, ValueType value, Span span=Span())</div><div class="ttdoc">Make a const value with certain data type. </div><div class="ttdef"><b>Definition:</b> op.h:954</div></div>
<div class="ttc" id="classtvm_1_1Range_html_a91e7301ca1d135ca5f8ed199efbb9818"><div class="ttname"><a href="classtvm_1_1Range.html#a91e7301ca1d135ca5f8ed199efbb9818">tvm::Range::FromMinExtent</a></div><div class="ttdeci">static Range FromMinExtent(PrimExpr min, PrimExpr extent, Span span=Span())</div><div class="ttdoc">construct a new range with min and extent The corresponding constructor is removed, because that is counter convention of tradition meaning of range(begin, end) </div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a3230e1735957c2045c89cf190e0f8c34"><div class="ttname"><a href="namespacetvm_1_1topi.html#a3230e1735957c2045c89cf190e0f8c34">tvm::topi::sliding_window</a></div><div class="ttdeci">Tensor sliding_window(const Tensor &x, int axis, Array< Integer > window_shape, Array< Integer > strides, std::string name="T_sliding_window", std::string tag="")</div><div class="ttdoc">Creates an operation to slide a window ove [...]
<div class="ttc" id="namespacetvm_html"><div class="ttname"><a href="namespacetvm.html">tvm</a></div><div class="ttdoc">runtime implementation for LibTorch/TorchScript. </div><div class="ttdef"><b>Definition:</b> analyzer.h:36</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_a9a4108863737f5022a539a45d8ad59f0"><div class="ttname"><a href="namespacetvm_1_1topi.html#a9a4108863737f5022a539a45d8ad59f0">tvm::topi::layout_transform</a></div><div class="ttdeci">Tensor layout_transform(const Tensor &src, const std::string &src_layout, const std::string &dst_layout, const std::string schedule_rule="None", const std::string name="T_layout_trans", const std::string tag=kInjective)</div><div c [...]
<div class="ttc" id="namespacetvm_1_1te_html"><div class="ttname"><a href="namespacetvm_1_1te.html">tvm::te</a></div><div class="ttdoc">Tensor expression language DSL. </div><div class="ttdef"><b>Definition:</b> extracted_task.h:33</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_abf2712c8265393c0582c9c7d5ae22da1"><div class="ttname"><a href="namespacetvm_1_1topi.html#abf2712c8265393c0582c9c7d5ae22da1">tvm::topi::tensordot</a></div><div class="ttdeci">Tensor tensordot(const Tensor &A, const tvm::te::Tensor &B, int axes=2, std::string name="T_tensordot", std::string tag=kMatMul)</div><div class="ttdoc">A generalization of matrix multiplication to tensors. </div><div class="ttdef"><b>Definition:</b> t [...]
<div class="ttc" id="namespacetvm_1_1topi_html_a2f4969306206bc345bd3be840f56999e"><div class="ttname"><a href="namespacetvm_1_1topi.html#a2f4969306206bc345bd3be840f56999e">tvm::topi::dynamic_strided_slice</a></div><div class="ttdeci">Tensor dynamic_strided_slice(const Tensor &x, const Array< PrimExpr > &begin, const Array< PrimExpr > &end, const Array< PrimExpr > &strides, std::string name="T_dynamic_strided_slice", std::string tag=kInjective)< [...]
<div class="ttc" id="namespacetvm_html_a5e4738caa6bcd0259af64b25e25dfd93"><div class="ttname"><a href="namespacetvm.html#a5e4738caa6bcd0259af64b25e25dfd93">tvm::ceil</a></div><div class="ttdeci">PrimExpr ceil(PrimExpr x, Span span=Span())</div><div class="ttdoc">Calculate ceil(x) </div></div>
<div class="ttc" id="classtvm_1_1tir_1_1Var_html"><div class="ttname"><a href="classtvm_1_1tir_1_1Var.html">tvm::tir::Var</a></div><div class="ttdoc">a named variable in TIR </div><div class="ttdef"><b>Definition:</b> var.h:88</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_af011847b6e7f72f1bec25eee05c80590"><div class="ttname"><a href="namespacetvm_1_1topi.html#af011847b6e7f72f1bec25eee05c80590">tvm::topi::where</a></div><div class="ttdeci">Tensor where(const Tensor &condition, const Tensor &x, const Tensor &y, std::string name="T_where", std::string tag=kBroadcast)</div><div class="ttdoc">Return the elements, either from x or y, depending on the condition. </div><div class="ttdef"><b>Def [...]
-<div class="ttc" id="namespacetvm_1_1topi_html_a3cf4e56cbd8144b9029672b7c5ebd161"><div class="ttname"><a href="namespacetvm_1_1topi.html#a3cf4e56cbd8144b9029672b7c5ebd161">tvm::topi::one_hot</a></div><div class="ttdeci">Tensor one_hot(const Tensor &indices, const PrimExpr on_value, const PrimExpr off_value, int depth, int axis, const DataType &dtype, Array< PrimExpr > oshape=Array< PrimExpr >(), const std::string name="T_one_hot", const std::string tag=kInje [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_a3cf4e56cbd8144b9029672b7c5ebd161"><div class="ttname"><a href="namespacetvm_1_1topi.html#a3cf4e56cbd8144b9029672b7c5ebd161">tvm::topi::one_hot</a></div><div class="ttdeci">Tensor one_hot(const Tensor &indices, const PrimExpr on_value, const PrimExpr off_value, int depth, int axis, const DataType &dtype, Array< PrimExpr > oshape=Array< PrimExpr >(), const std::string name="T_one_hot", const std::string tag=kInje [...]
<div class="ttc" id="namespacetvm_html_a353217978feabae3575560bf1586885f"><div class="ttname"><a href="namespacetvm.html#a353217978feabae3575560bf1586885f">tvm::if_then_else</a></div><div class="ttdeci">PrimExpr if_then_else(PrimExpr cond, PrimExpr true_value, PrimExpr false_value, Span span=Span())</div><div class="ttdoc">Conditional expression. </div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_aead477c6c9d4f4589d22b8acff82040c"><div class="ttname"><a href="namespacetvm_1_1topi.html#aead477c6c9d4f4589d22b8acff82040c">tvm::topi::matrix_set_diag</a></div><div class="ttdeci">Tensor matrix_set_diag(const Tensor &input, const Tensor &diagonal, int k1, int k2, bool super_diag_right_align, bool sub_diag_right_align, const std::string name="T_matrix_set_diag", const std::string tag=kInjective)</div><div class="ttdoc">Retu [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_aead477c6c9d4f4589d22b8acff82040c"><div class="ttname"><a href="namespacetvm_1_1topi.html#aead477c6c9d4f4589d22b8acff82040c">tvm::topi::matrix_set_diag</a></div><div class="ttdeci">Tensor matrix_set_diag(const Tensor &input, const Tensor &diagonal, int k1, int k2, bool super_diag_right_align, bool sub_diag_right_align, const std::string name="T_matrix_set_diag", const std::string tag=kInjective)</div><div class="ttdoc">Retu [...]
<div class="ttc" id="namespacetvm_1_1topi_html_aa9f70e7392653f38da6ff53f5f554f98"><div class="ttname"><a href="namespacetvm_1_1topi.html#aa9f70e7392653f38da6ff53f5f554f98">tvm::topi::kMatMul</a></div><div class="ttdeci">constexpr auto kMatMul</div><div class="ttdef"><b>Definition:</b> tags.h:37</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a29e22aa45900dad3b6f9f705bb1dc688"><div class="ttname"><a href="namespacetvm_1_1topi.html#a29e22aa45900dad3b6f9f705bb1dc688">tvm::topi::kInjective</a></div><div class="ttdeci">constexpr auto kInjective</div><div class="ttdef"><b>Definition:</b> tags.h:33</div></div>
<div class="ttc" id="classtvm_1_1arith_1_1Analyzer_html_a9b440f852f12ad0a4d8ed5ed97054425"><div class="ttname"><a href="classtvm_1_1arith_1_1Analyzer.html#a9b440f852f12ad0a4d8ed5ed97054425">tvm::arith::Analyzer::Simplify</a></div><div class="ttdeci">PrimExpr Simplify(const PrimExpr &expr, int steps=2)</div><div class="ttdoc">Simplify expr. </div></div>
@@ -94,12 +95,11 @@ $(function() {
<div class="ttc" id="namespacetvm_1_1topi_html_a8811a02532bbe3047986bf1a8449ac0e"><div class="ttname"><a href="namespacetvm_1_1topi.html#a8811a02532bbe3047986bf1a8449ac0e">tvm::topi::unravel_index</a></div><div class="ttdeci">Tensor unravel_index(const Tensor &x, const Tensor &shape, std::string name="T_unravel", std::string tag=kInjective)</div><div class="ttdoc">Converts a flat index or array of flat indices into a tuple of coordinate arrays. ...</div><div class="ttde [...]
<div class="ttc" id="namespacetvm_1_1topi_html_af4e59b01a5842baf6b47ad3f83731f53"><div class="ttname"><a href="namespacetvm_1_1topi.html#af4e59b01a5842baf6b47ad3f83731f53">tvm::topi::split</a></div><div class="ttdeci">Array< Tensor > split(const Tensor &x, Array< PrimExpr > split_indices, int axis, std::string name="T_split", std::string tag=kInjective)</div><div class="ttdoc">Split a tensor into multiple sub-tensors. </div><div class="ttdef"><b>Definition:</b> [...]
<div class="ttc" id="classtvm_1_1PrimExpr_html_a18a8a0fef87b74b18fc254f49465cc56"><div class="ttname"><a href="classtvm_1_1PrimExpr.html#a18a8a0fef87b74b18fc254f49465cc56">tvm::PrimExpr::dtype</a></div><div class="ttdeci">DataType dtype() const</div><div class="ttdef"><b>Definition:</b> expr.h:128</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a6fec3a88d419cd018cf1041cf3d99204"><div class="ttname"><a href="namespacetvm_1_1topi.html#a6fec3a88d419cd018cf1041cf3d99204">tvm::topi::parse_auto_scheduler_layout</a></div><div class="ttdeci">void parse_auto_scheduler_layout(const String &layout, Array< PrimExpr > *shape, std::vector< std::string > *axes)</div><div class="ttdoc">Utility function for auto_scheduler_layout_transform. </div><div class="ttdef"><b>Definition:</b> [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_a6fec3a88d419cd018cf1041cf3d99204"><div class="ttname"><a href="namespacetvm_1_1topi.html#a6fec3a88d419cd018cf1041cf3d99204">tvm::topi::parse_auto_scheduler_layout</a></div><div class="ttdeci">void parse_auto_scheduler_layout(const String &layout, Array< PrimExpr > *shape, std::vector< std::string > *axes)</div><div class="ttdoc">Utility function for auto_scheduler_layout_transform. </div><div class="ttdef"><b>Definition:</b> [...]
<div class="ttc" id="namespacetvm_1_1topi_html_af392bc8e9dc0d34792652a4c0cf92243"><div class="ttname"><a href="namespacetvm_1_1topi.html#af392bc8e9dc0d34792652a4c0cf92243">tvm::topi::gather</a></div><div class="ttdeci">Tensor gather(const Tensor &data, int axis, const Tensor &indices, std::string name="T_gather", std::string tag=kInjective)</div><div class="ttdoc">Gather values along given axis from given indices. </div><div class="ttdef"><b>Definition:</b> transform.h: [...]
<div class="ttc" id="classtvm_1_1te_1_1Tensor_html_a7c8d978e9905bb0f0f88226d4f5cbe7a"><div class="ttname"><a href="classtvm_1_1te_1_1Tensor.html#a7c8d978e9905bb0f0f88226d4f5cbe7a">tvm::te::Tensor::ndim</a></div><div class="ttdeci">size_t ndim() const</div><div class="ttdef"><b>Definition:</b> tensor.h:214</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a5efc5887d91eb272e31c10f4e014a034"><div class="ttname"><a href="namespacetvm_1_1topi.html#a5efc5887d91eb272e31c10f4e014a034">tvm::topi::dyn_tile</a></div><div class="ttdeci">Tensor dyn_tile(const Tensor &x, Array< PrimExpr > new_shape, size_t rdim, std::string name="T_tile", std::string tag=kBroadcast)</div><div class="ttdoc">Creates an operation to tile elements of an array. </div><div class="ttdef"><b>Definition:</b> tr [...]
-<div class="ttc" id="namespacetvm_1_1topi_html_a8a41a08eee70607889b738946ed97873"><div class="ttname"><a href="namespacetvm_1_1topi.html#a8a41a08eee70607889b738946ed97873">tvm::topi::layout_transform</a></div><div class="ttdeci">Tensor layout_transform(const Tensor &src, const std::string &src_layout, const std::string &dst_layout, const std::string name="T_layout_trans", const std::string tag=kInjective)</div><div class="ttdoc">Transform the layout according to src [...]
-<div class="ttc" id="namespacetvm_1_1topi_html_a8e10f74deef4f22a9dc4b0a0b4370b08"><div class="ttname"><a href="namespacetvm_1_1topi.html#a8e10f74deef4f22a9dc4b0a0b4370b08">tvm::topi::auto_scheduler_layout_transform</a></div><div class="ttdeci">Tensor auto_scheduler_layout_transform(const Tensor &src, const String &src_layout, const String &dst_layout, const String name="T_auto_scheduler_layout_trans", const String tag=kInjective)</div><div class="ttdoc">Transform th [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_a8e10f74deef4f22a9dc4b0a0b4370b08"><div class="ttname"><a href="namespacetvm_1_1topi.html#a8e10f74deef4f22a9dc4b0a0b4370b08">tvm::topi::auto_scheduler_layout_transform</a></div><div class="ttdeci">Tensor auto_scheduler_layout_transform(const Tensor &src, const String &src_layout, const String &dst_layout, const String name="T_auto_scheduler_layout_trans", const String tag=kInjective)</div><div class="ttdoc">Transform th [...]
<div class="ttc" id="namespacetvm_html_a4bfb789a86d95f6241b50fd26f269c28"><div class="ttname"><a href="namespacetvm.html#a4bfb789a86d95f6241b50fd26f269c28">tvm::cast</a></div><div class="ttdeci">PrimExpr cast(const DataType &t, PrimExpr value, Span span=Span())</div><div class="ttdoc">cast value to type. </div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a845e38c0f34017d45ec318935b6ddf17"><div class="ttname"><a href="namespacetvm_1_1topi.html#a845e38c0f34017d45ec318935b6ddf17">tvm::topi::squeeze</a></div><div class="ttdeci">Tensor squeeze(const Tensor &x, Array< Integer > axis, bool atleast1d=false, std::string name="T_squeeze", std::string tag=kInjective)</div><div class="ttdoc">Remove size 1 dimensions from the shape of a tensor. The removed dimensions must have a const [...]
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html_ab5db2ee9a8be71931324dac552be24c4"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html#ab5db2ee9a8be71931324dac552be24c4">tvm::runtime::Array::Set</a></div><div class="ttdeci">void Set(int64_t i, T value)</div><div class="ttdoc">set i-th element of the array. </div><div class="ttdef"><b>Definition:</b> array.h:591</div></div>
@@ -130,7 +130,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1IntImmNode_html_a81f4c116ffb5931fdd64639eacad415d"><div class="ttname"><a href="classtvm_1_1IntImmNode.html#a81f4c116ffb5931fdd64639eacad415d">tvm::IntImmNode::value</a></div><div class="ttdeci">int64_t value</div><div class="ttdoc">the Internal value. </div><div class="ttdef"><b>Definition:</b> expr.h:494</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1String_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1String.html">tvm::runtime::String</a></div><div class="ttdoc">Reference to string objects. </div><div class="ttdef"><b>Definition:</b> string.h:98</div></div>
<div class="ttc" id="namespacetvm_1_1tir_html_aed3f57cf8d1c3546f075701898c5b70f"><div class="ttname"><a href="namespacetvm_1_1tir.html#aed3f57cf8d1c3546f075701898c5b70f">tvm::tir::make_zero</a></div><div class="ttdeci">PrimExpr make_zero(DataType t, Span span=Span())</div><div class="ttdoc">Make a const zero expr. </div><div class="ttdef"><b>Definition:</b> op.h:962</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1760</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_af30c02f3a3f37c7963b3af60fb9c72a1"><div class="ttname"><a href="namespacetvm_1_1topi.html#af30c02f3a3f37c7963b3af60fb9c72a1">tvm::topi::shape</a></div><div class="ttdeci">Tensor shape(const Tensor &src, DataType dtype, const std::string name="T_shape", const std::string tag=kInjective)</div><div class="ttdoc">Get the shape of input tensor. </div><div class="ttdef"><b>Definition:</b> transform.h:1768</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a453f9f5a35a9852a086ce0228ee22164"><div class="ttname"><a href="namespacetvm_1_1topi.html#a453f9f5a35a9852a086ce0228ee22164">tvm::topi::meshgrid</a></div><div class="ttdeci">Array< Tensor > meshgrid(const Array< Tensor > &inputs, const std::string &indexing, std::string name="T_meshgrid", std::string tag=kInjective)</div><div class="ttdoc">Produce grids by expanding input over dimensions defined by other inputs. [...]
<div class="ttc" id="namespacetvm_1_1te_html_aae384e9b73c2271905486e4a74b69265"><div class="ttname"><a href="namespacetvm_1_1te.html#aae384e9b73c2271905486e4a74b69265">tvm::te::reduce_axis</a></div><div class="ttdeci">IterVar reduce_axis(Range dom, std::string name="rv")</div><div class="ttdoc">Create a new IterVar for reduction operations. </div></div>
<div class="ttc" id="namespacetvm_html_a2428ea0e23bd9f7218aebd066bb2cd88"><div class="ttname"><a href="namespacetvm.html#a2428ea0e23bd9f7218aebd066bb2cd88">tvm::truncmod</a></div><div class="ttdeci">PrimExpr truncmod(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">compute the remainder of truncdiv </div></div>
@@ -139,15 +139,16 @@ $(function() {
<div class="ttc" id="classtvm_1_1te_1_1Tensor_html"><div class="ttname"><a href="classtvm_1_1te_1_1Tensor.html">tvm::te::Tensor</a></div><div class="ttdoc">Tensor structure representing a possible input, or intermediate computation result. </div><div class="ttdef"><b>Definition:</b> tensor.h:102</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html_ae9734c8e6324f9be27fce8d45d0aa6f4"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html#ae9734c8e6324f9be27fce8d45d0aa6f4">tvm::runtime::Array::begin</a></div><div class="ttdeci">iterator begin() const</div><div class="ttdef"><b>Definition:</b> array.h:387</div></div>
<div class="ttc" id="operation_8h_html"><div class="ttname"><a href="operation_8h.html">operation.h</a></div><div class="ttdoc">Operation node can generate one or multiple Tensors. </div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a3df09fe365656f953c8145a8cf0b3fa3"><div class="ttname"><a href="namespacetvm_1_1topi.html#a3df09fe365656f953c8145a8cf0b3fa3">tvm::topi::meta_schedule_layout_transform</a></div><div class="ttdeci">Tensor meta_schedule_layout_transform(const Tensor &src, const tir::IndexMap &index_map, const String name="T_meta_schedule_layout_trans", const String tag=kInjective)</div><div class="ttdoc">Transform the meta-schedule generated l [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_a3df09fe365656f953c8145a8cf0b3fa3"><div class="ttname"><a href="namespacetvm_1_1topi.html#a3df09fe365656f953c8145a8cf0b3fa3">tvm::topi::meta_schedule_layout_transform</a></div><div class="ttdeci">Tensor meta_schedule_layout_transform(const Tensor &src, const tir::IndexMap &index_map, const String name="T_meta_schedule_layout_trans", const String tag=kInjective)</div><div class="ttdoc">Transform the meta-schedule generated l [...]
<div class="ttc" id="classtvm_1_1tir_1_1Select_html"><div class="ttname"><a href="classtvm_1_1tir_1_1Select.html">tvm::tir::Select</a></div><div class="ttdoc">Managed reference to SelectNode. </div><div class="ttdef"><b>Definition:</b> expr.h:609</div></div>
<div class="ttc" id="classtvm_1_1tir_1_1BijectiveLayout_html"><div class="ttname"><a href="classtvm_1_1tir_1_1BijectiveLayout.html">tvm::tir::BijectiveLayout</a></div><div class="ttdoc">Bijective function mapping for data layout transformation. Given two Layout, BijectiveLayout build an...</div><div class="ttdef"><b>Definition:</b> data_layout.h:332</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a1488ee98fd053e8b01b481f720df77fa"><div class="ttname"><a href="namespacetvm_1_1topi.html#a1488ee98fd053e8b01b481f720df77fa">tvm::topi::transpose</a></div><div class="ttdeci">Tensor transpose(const Tensor &x, Array< Integer > axes, std::string name="T_transpose", std::string tag=kInjective)</div><div class="ttdoc">Permute the dimensions of an array. </div><div class="ttdef"><b>Definition:</b> transform.h:196</div></div>
<div class="ttc" id="namespacetvm_html_a4f1398024c0af23699447ef910b654b8"><div class="ttname"><a href="namespacetvm.html#a4f1398024c0af23699447ef910b654b8">tvm::max_value</a></div><div class="ttdeci">PrimExpr max_value(const DataType &dtype, Span span=Span())</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a92d2a9f409bae6731c5e5247de0b8c48"><div class="ttname"><a href="namespacetvm_1_1topi.html#a92d2a9f409bae6731c5e5247de0b8c48">tvm::topi::ndarray_size</a></div><div class="ttdeci">Tensor ndarray_size(const Tensor &src, const DataType &dtype, const std::string &name="ndarray_size", const std::string &tag=kInjective)</div><div class="ttdoc">Get the size of input tensor. </div><div class="ttdef"><b>Definition:</b> transf [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_a92d2a9f409bae6731c5e5247de0b8c48"><div class="ttname"><a href="namespacetvm_1_1topi.html#a92d2a9f409bae6731c5e5247de0b8c48">tvm::topi::ndarray_size</a></div><div class="ttdeci">Tensor ndarray_size(const Tensor &src, const DataType &dtype, const std::string &name="ndarray_size", const std::string &tag=kInjective)</div><div class="ttdoc">Get the size of input tensor. </div><div class="ttdef"><b>Definition:</b> transf [...]
<div class="ttc" id="namespacetvm_1_1topi_html_a037a112cc5c556107797e36e7feb0873"><div class="ttname"><a href="namespacetvm_1_1topi.html#a037a112cc5c556107797e36e7feb0873">tvm::topi::sequence_mask</a></div><div class="ttdeci">Tensor sequence_mask(const Tensor &data, const Tensor &valid_length, double mask_value, int axis, std::string name="T_sequence_mask", std::string tag=kInjective)</div><div class="ttdoc">Mask the out-of-boundary elements of each sequence. </div><div [...]
+<div class="ttc" id="classtvm_1_1runtime_1_1Map_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1Map.html">tvm::runtime::Map</a></div><div class="ttdoc">Map container of NodeRef->NodeRef in DSL graph. Map implements copy on write semantics, which means map is mutable but copy will happen when array is referenced in more than two places. </div><div class="ttdef"><b>Definition:</b> map.h:1271</div></div>
<div class="ttc" id="namespacetvm_html_a0da40d3e210aa3b38a17982a7b7866b8"><div class="ttname"><a href="namespacetvm.html#a0da40d3e210aa3b38a17982a7b7866b8">tvm::ret</a></div><div class="ttdeci">PrimExpr ret(PrimExpr value, Span span=Span())</div><div class="ttdoc">Return the value. </div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a6d9189f6ceb05cf0a309dbe3f2730b16"><div class="ttname"><a href="namespacetvm_1_1topi.html#a6d9189f6ceb05cf0a309dbe3f2730b16">tvm::topi::adv_index</a></div><div class="ttdeci">Tensor adv_index(const Tensor &data, const Array< Tensor > &indices, const std::string name="advanced_index", const std::string tag=kInjective)</div><div class="ttdoc">Numpy style advanced indexing with tensor. </div><div class="ttdef"><b>Definit [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_a6d9189f6ceb05cf0a309dbe3f2730b16"><div class="ttname"><a href="namespacetvm_1_1topi.html#a6d9189f6ceb05cf0a309dbe3f2730b16">tvm::topi::adv_index</a></div><div class="ttdeci">Tensor adv_index(const Tensor &data, const Array< Tensor > &indices, const std::string name="advanced_index", const std::string tag=kInjective)</div><div class="ttdoc">Numpy style advanced indexing with tensor. </div><div class="ttdef"><b>Definit [...]
<div class="ttc" id="tags_8h_html"><div class="ttname"><a href="tags_8h.html">tags.h</a></div><div class="ttdoc">External function interface to rocBLAS libraries. </div></div>
<div class="ttc" id="namespacetvm_1_1te_html_afe4f57aeb3dd5ae9c0b58135e14d67ca"><div class="ttname"><a href="namespacetvm_1_1te.html#afe4f57aeb3dd5ae9c0b58135e14d67ca">tvm::te::compute</a></div><div class="ttdeci">Tensor compute(Array< PrimExpr > shape, FCompute fcompute, std::string name="tensor", std::string tag="", Map< String, ObjectRef > attrs={})</div><div class="ttdoc">Construct a new tensor by computing over shape, using the computation rule: resul [...]
<div class="ttc" id="namespacetvm_1_1topi_html_ab8ad5eed3079de21c92a7639ed370096"><div class="ttname"><a href="namespacetvm_1_1topi.html#ab8ad5eed3079de21c92a7639ed370096">tvm::topi::reverse_sequence</a></div><div class="ttdeci">Tensor reverse_sequence(const Tensor &x, const Tensor &seq_lengths, int seq_axis=1, int batch_axis=0, std::string name="T_reverse_sequence", std::string tag=kInjective)</div><div class="ttdoc">Reverse the tensor for variable length slices. Input [...]
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index 7a6c1fdac2..df029742a0 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1621,7 +1621,7 @@ history states as starting point to perform Evolutionary Search).</p></li>
<dl class="py class">
<dt class="sig sig-object py" id="tvm.auto_scheduler.SketchPolicy">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
<dd><p>The search policy that searches in a hierarchical search space defined by sketches.
The policy randomly samples programs from the space defined by sketches and use evolutionary
search to fine-tune them.</p>
@@ -1905,7 +1905,7 @@ Candidates:
<dl class="py function">
<dt class="sig sig-object py" id="tvm.auto_scheduler.auto_schedule">
-<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
<dd><p>THIS API IS DEPRECATED.</p>
<p>Run auto scheduling search for a task.</p>
<dl class="field-list simple">
diff --git a/docs/reference/api/python/topi.html b/docs/reference/api/python/topi.html
index 67542c52db..d23a758441 100644
--- a/docs/reference/api/python/topi.html
+++ b/docs/reference/api/python/topi.html
@@ -2484,7 +2484,7 @@ Otherwise returns Tensor.</p>
<dl class="py function">
<dt class="sig sig-object py">
-<span class="sig-prename descclassname"><span class="pre">tvm.topi.</span></span><span class="sig-name descname"><span class="pre">layout_transform</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">array</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">src_layout</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dst_layout</span></span></em><span class="sig-paren">)</span></dt>
+<span class="sig-prename descclassname"><span class="pre">tvm.topi.</span></span><span class="sig-name descname"><span class="pre">layout_transform</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">array</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">src_layout</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dst_layout</span></span></em>, <em class="sig-param"><span class="n"><span [...]
<dd><p>Transform the layout according to src_layout and dst_layout</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
@@ -2492,6 +2492,7 @@ Otherwise returns Tensor.</p>
<li><p><strong>array</strong> (<a class="reference internal" href="te.html#tvm.te.Tensor" title="tvm.te.Tensor"><em>tvm.te.Tensor</em></a>) – The source array.</p></li>
<li><p><strong>src_layout</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.11)"><em>str</em></a>) – the source layout.</p></li>
<li><p><strong>dst_layout</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.11)"><em>str</em></a>) – the destination layout.</p></li>
+<li><p><strong>schedule_rule</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.11)"><em>str</em></a>) – the schedule rule to apply if any</p></li>
</ul>
</dd>
</dl>
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index 4cea2d23d7..c27e6853d6 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/3f56a95b8/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -141,7 +141,7 @@
<div class="tsd-signature tsd-kind-icon">bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Uint8Array</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
</ul>
</aside>
</section>
@@ -151,7 +151,7 @@
<div class="tsd-signature tsd-kind-icon">offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
</ul>
</aside>
</section>
@@ -168,7 +168,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">Uint8Array</span></h4>
@@ -185,7 +185,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -202,7 +202,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/cachedcallstack.html b/docs/reference/api/typedoc/classes/cachedcallstack.html
index 564f98c7c0..d71eb784e7 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/3f56a95b8/web/src/memory.ts#L223">memory.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L208">memory.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L312">memory.ts:312</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L284">memory.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L388">memory.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L376">memory.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L267">memory.ts:267</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L243">memory.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L321">memory.ts:321</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L252">memory.ts:252</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L359">memory.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L342">memory.ts:342</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L350">memory.ts:350</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L326">memory.ts:326</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L363">memory.ts:363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L346">memory.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/memory.ts#L334">memory.ts:334</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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 2bf17f2610..323467d8f4 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/3f56a95b8/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L359">runtime.ts:359</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
<div class="tsd-signature tsd-kind-icon">bits<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L357">runtime.ts:357</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L357">runtime.ts:357</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">code<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L355">runtime.ts:355</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -177,7 +177,7 @@
<div class="tsd-signature tsd-kind-icon">lanes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L359">runtime.ts:359</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -199,7 +199,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L376">runtime.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L376">runtime.ts:376</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -216,7 +216,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L367">runtime.ts:367</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L367">runtime.ts:367</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/dldevice.html b/docs/reference/api/typedoc/classes/dldevice.html
index cc700eada7..a8ebf5eb7b 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/3f56a95b8/web/src/runtime.ts#L299">runtime.ts:299</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L299">runtime.ts:299</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L297">runtime.ts:297</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -161,7 +161,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L295">runtime.ts:295</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -183,7 +183,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L320">runtime.ts:320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L320">runtime.ts:320</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -205,7 +205,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L327">runtime.ts:327</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L327">runtime.ts:327</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/environment.html b/docs/reference/api/typedoc/classes/environment.html
index 92d284228a..5c5453e8df 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/3f56a95b8/web/src/environment.ts#L86">environment.ts:86</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/environment.ts#L70">environment.ts:70</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/environment.ts#L69">environment.ts:69</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/environment.ts#L78">environment.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/environment.ts#L84">environment.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/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/3f56a95b8/web/src/environment.ts#L105">environment.ts:105</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/environment.ts#L105">environment.ts:105</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ffilibrary.html b/docs/reference/api/typedoc/classes/ffilibrary.html
index 0a88772c14..37eaa23b8c 100644
--- a/docs/reference/api/typedoc/classes/ffilibrary.html
+++ b/docs/reference/api/typedoc/classes/ffilibrary.html
@@ -131,7 +131,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L50">runtime.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L50">runtime.ts:50</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -156,7 +156,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L47">runtime.ts:47</a></li>
</ul>
</aside>
</section>
@@ -166,7 +166,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L46">runtime.ts:46</a></li>
</ul>
</aside>
</section>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L45">runtime.ts:45</a></li>
</ul>
</aside>
</section>
@@ -186,7 +186,7 @@
<div class="tsd-signature tsd-kind-icon">webGPUContext<span class="tsd-signature-symbol">:</span> <a href="webgpucontext.html" class="tsd-signature-type">WebGPUContext</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L48">runtime.ts:48</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L48">runtime.ts:48</a></li>
</ul>
</aside>
</section>
@@ -203,7 +203,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L77">runtime.ts:77</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L77">runtime.ts:77</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -226,7 +226,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L67">runtime.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L67">runtime.ts:67</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -243,7 +243,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L85">runtime.ts:85</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L85">runtime.ts:85</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <a href="cachedcallstack.html" class="tsd-signature-type">CachedCallStack</a></h4>
@@ -260,7 +260,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L96">runtime.ts:96</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L96">runtime.ts:96</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -283,7 +283,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L73">runtime.ts:73</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L73">runtime.ts:73</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/instance.html b/docs/reference/api/typedoc/classes/instance.html
index 27d7b37f0e..d28e41e6e8 100644
--- a/docs/reference/api/typedoc/classes/instance.html
+++ b/docs/reference/api/typedoc/classes/instance.html
@@ -161,7 +161,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L844">runtime.ts:844</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L844">runtime.ts:844</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L834">runtime.ts:834</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L834">runtime.ts:834</a></li>
</ul>
</aside>
</section>
@@ -234,7 +234,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L833">runtime.ts:833</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L833">runtime.ts:833</a></li>
</ul>
</aside>
</section>
@@ -251,7 +251,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L973">runtime.ts:973</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L973">runtime.ts:973</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -296,7 +296,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L932">runtime.ts:932</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -318,7 +318,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L901">runtime.ts:901</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L901">runtime.ts:901</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -381,7 +381,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -412,7 +412,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -453,7 +453,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -491,7 +491,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L922">runtime.ts:922</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L922">runtime.ts:922</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -508,7 +508,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -552,7 +552,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L943">runtime.ts:943</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L943">runtime.ts:943</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -577,7 +577,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -609,7 +609,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -640,7 +640,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -672,7 +672,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -695,7 +695,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -729,7 +729,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L986">runtime.ts:986</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L986">runtime.ts:986</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -769,7 +769,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -817,7 +817,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -857,7 +857,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -900,7 +900,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -938,7 +938,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -990,7 +990,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1014,7 +1014,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1046,7 +1046,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1078,7 +1078,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1110,7 +1110,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1141,7 +1141,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L957">runtime.ts:957</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L957">runtime.ts:957</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/memory.html b/docs/reference/api/typedoc/classes/memory.html
index c9a1d07c83..f9a6c26d7d 100644
--- a/docs/reference/api/typedoc/classes/memory.html
+++ b/docs/reference/api/typedoc/classes/memory.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L40">memory.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L40">memory.ts:40</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -152,7 +152,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Memory</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L32">memory.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L32">memory.ts:32</a></li>
</ul>
</aside>
</section>
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span><span class="tsd-signature-symbol"> = true</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L33">memory.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L33">memory.ts:33</a></li>
</ul>
</aside>
</section>
@@ -179,7 +179,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L154">memory.ts:154</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L154">memory.ts:154</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -210,7 +210,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L90">memory.ts:90</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L90">memory.ts:90</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -233,7 +233,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L97">memory.ts:97</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L97">memory.ts:97</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -256,7 +256,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L74">memory.ts:74</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L74">memory.ts:74</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -279,7 +279,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L81">memory.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L81">memory.ts:81</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -302,7 +302,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L104">memory.ts:104</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L104">memory.ts:104</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -325,7 +325,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L132">memory.ts:132</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L132">memory.ts:132</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -362,7 +362,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L145">memory.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L145">memory.ts:145</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -393,7 +393,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L60">memory.ts:60</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L60">memory.ts:60</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -416,7 +416,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L67">memory.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L67">memory.ts:67</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -439,7 +439,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L53">memory.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L53">memory.ts:53</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -462,7 +462,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L114">memory.ts:114</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L114">memory.ts:114</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -485,7 +485,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L124">memory.ts:124</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L124">memory.ts:124</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -502,7 +502,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/memory.ts#L175">memory.ts:175</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/memory.ts#L175">memory.ts:175</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/module.html b/docs/reference/api/typedoc/classes/module.html
index 8a00d1e570..a199636e8c 100644
--- a/docs/reference/api/typedoc/classes/module.html
+++ b/docs/reference/api/typedoc/classes/module.html
@@ -119,7 +119,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L614">runtime.ts:614</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L614">runtime.ts:614</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -169,7 +169,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L626">runtime.ts:626</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L626">runtime.ts:626</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -186,7 +186,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L653">runtime.ts:653</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L653">runtime.ts:653</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -218,7 +218,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L641">runtime.ts:641</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L641">runtime.ts:641</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -250,7 +250,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L687">runtime.ts:687</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L687">runtime.ts:687</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ndarray.html b/docs/reference/api/typedoc/classes/ndarray.html
index 5fb287cdda..01c2063cd6 100644
--- a/docs/reference/api/typedoc/classes/ndarray.html
+++ b/docs/reference/api/typedoc/classes/ndarray.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L401">runtime.ts:401</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L401">runtime.ts:401</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <a href="dldevice.html" class="tsd-signature-type">DLDevice</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L394">runtime.ts:394</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L394">runtime.ts:394</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -173,7 +173,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L390">runtime.ts:390</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L390">runtime.ts:390</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -188,7 +188,7 @@
<div class="tsd-signature tsd-kind-icon">ndim<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L388">runtime.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L388">runtime.ts:388</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -203,7 +203,7 @@
<div class="tsd-signature tsd-kind-icon">shape<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L392">runtime.ts:392</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L392">runtime.ts:392</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -225,7 +225,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L480">runtime.ts:480</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L480">runtime.ts:480</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -258,7 +258,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L524">runtime.ts:524</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L524">runtime.ts:524</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -290,7 +290,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L465">runtime.ts:465</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L465">runtime.ts:465</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -307,7 +307,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L458">runtime.ts:458</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L458">runtime.ts:458</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -339,7 +339,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L584">runtime.ts:584</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L584">runtime.ts:584</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -363,7 +363,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L553">runtime.ts:553</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L553">runtime.ts:553</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/packedfunccell.html b/docs/reference/api/typedoc/classes/packedfunccell.html
index 3645dd6599..db3fd2f1c6 100644
--- a/docs/reference/api/typedoc/classes/packedfunccell.html
+++ b/docs/reference/api/typedoc/classes/packedfunccell.html
@@ -117,7 +117,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L248">runtime.ts:248</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L248">runtime.ts:248</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L255">runtime.ts:255</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L255">runtime.ts:255</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -163,7 +163,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L264">runtime.ts:264</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L264">runtime.ts:264</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/rpcserver.html b/docs/reference/api/typedoc/classes/rpcserver.html
index 9bf5ad73e6..53ee629527 100644
--- a/docs/reference/api/typedoc/classes/rpcserver.html
+++ b/docs/reference/api/typedoc/classes/rpcserver.html
@@ -115,7 +115,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">get<wbr>Imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">unknown</span><span class="tsd-signat [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -201,7 +201,7 @@
<div class="tsd-signature tsd-kind-icon">key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
</ul>
</aside>
</section>
@@ -211,7 +211,7 @@
<div class="tsd-signature tsd-kind-icon">logger<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>msg<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -242,7 +242,7 @@
<div class="tsd-signature tsd-kind-icon">socket<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">WebSocket</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
</ul>
</aside>
</section>
@@ -252,7 +252,7 @@
<div class="tsd-signature tsd-kind-icon">state<span class="tsd-signature-symbol">:</span> <a href="../enums/rpcserverstate.html" class="tsd-signature-type">RPCServerState</a><span class="tsd-signature-symbol"> = RPCServerState.InitHeader</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
</ul>
</aside>
</section>
@@ -262,7 +262,7 @@
<div class="tsd-signature tsd-kind-icon">url<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/classes/runtimecontext.html b/docs/reference/api/typedoc/classes/runtimecontext.html
index e8afa7a73d..2a5eecdbf2 100644
--- a/docs/reference/api/typedoc/classes/runtimecontext.html
+++ b/docs/reference/api/typedoc/classes/runtimecontext.html
@@ -132,7 +132,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L148">runtime.ts:148</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L148">runtime.ts:148</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -172,7 +172,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Item<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L143">runtime.ts:143</a></li>
</ul>
</aside>
</section>
@@ -182,7 +182,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Size<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L144">runtime.ts:144</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L144">runtime.ts:144</a></li>
</ul>
</aside>
</section>
@@ -192,7 +192,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Make<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L145">runtime.ts:145</a></li>
</ul>
</aside>
</section>
@@ -202,7 +202,7 @@
<div class="tsd-signature tsd-kind-icon">get<wbr>Sys<wbr>Lib<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L146">runtime.ts:146</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L146">runtime.ts:146</a></li>
</ul>
</aside>
</section>
@@ -219,7 +219,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L189">runtime.ts:189</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L189">runtime.ts:189</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -263,7 +263,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L163">runtime.ts:163</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L163">runtime.ts:163</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -280,7 +280,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L208">runtime.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L208">runtime.ts:208</a></li>
</ul>
</aside>
<h4 class="tsd-type-parameters-title">Type parameters</h4>
@@ -309,7 +309,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L157">runtime.ts:157</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -326,7 +326,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L167">runtime.ts:167</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L167">runtime.ts:167</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -343,7 +343,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L198">runtime.ts:198</a></li>
</ul>
</aside>
<h4 class="tsd-type-parameters-title">Type parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/scalar.html b/docs/reference/api/typedoc/classes/scalar.html
index e7dd387b0f..06f9ca9ec1 100644
--- a/docs/reference/api/typedoc/classes/scalar.html
+++ b/docs/reference/api/typedoc/classes/scalar.html
@@ -112,7 +112,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L235">runtime.ts:235</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -137,7 +137,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L235">runtime.ts:235</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -152,7 +152,7 @@
<div class="tsd-signature tsd-kind-icon">value<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L233">runtime.ts:233</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L233">runtime.ts:233</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmarray.html b/docs/reference/api/typedoc/classes/tvmarray.html
index 63d5637e5d..0f2dd6df11 100644
--- a/docs/reference/api/typedoc/classes/tvmarray.html
+++ b/docs/reference/api/typedoc/classes/tvmarray.html
@@ -133,7 +133,7 @@
<aside class="tsd-sources">
<p>Overrides <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#constructor">constructor</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L784">runtime.ts:784</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L784">runtime.ts:784</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -162,7 +162,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#ctx">ctx</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
</section>
@@ -180,7 +180,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#dispose">dispose</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L715">runtime.ts:715</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -197,7 +197,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L804">runtime.ts:804</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L804">runtime.ts:804</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -230,7 +230,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#gethandle">getHandle</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L730">runtime.ts:730</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -262,7 +262,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L796">runtime.ts:796</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L796">runtime.ts:796</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -283,7 +283,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typeindex">typeIndex</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L738">runtime.ts:738</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -306,7 +306,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typekey">typeKey</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L758">runtime.ts:758</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmobject.html b/docs/reference/api/typedoc/classes/tvmobject.html
index 5fca53a6de..af9e181be1 100644
--- a/docs/reference/api/typedoc/classes/tvmobject.html
+++ b/docs/reference/api/typedoc/classes/tvmobject.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
<div class="tsd-signature tsd-kind-icon">ctx<span class="tsd-signature-symbol">:</span> <a href="runtimecontext.html" class="tsd-signature-type">RuntimeContext</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
</section>
@@ -175,7 +175,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L715">runtime.ts:715</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -192,7 +192,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L730">runtime.ts:730</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L738">runtime.ts:738</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -246,7 +246,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L758">runtime.ts:758</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/webgpucontext.html b/docs/reference/api/typedoc/classes/webgpucontext.html
index b0c9179d3d..6914b6e609 100644
--- a/docs/reference/api/typedoc/classes/webgpucontext.html
+++ b/docs/reference/api/typedoc/classes/webgpucontext.html
@@ -120,7 +120,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -145,7 +145,7 @@
<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">GPUDevice</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
</ul>
</aside>
</section>
@@ -155,7 +155,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
</ul>
</aside>
</section>
@@ -172,7 +172,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -209,7 +209,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/enums/argtypecode.html b/docs/reference/api/typedoc/enums/argtypecode.html
index df8ed1f8e5..70a4dd6fd3 100644
--- a/docs/reference/api/typedoc/enums/argtypecode.html
+++ b/docs/reference/api/typedoc/enums/argtypecode.html
@@ -106,7 +106,7 @@
<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 6</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L242">ctypes.ts:242</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L242">ctypes.ts:242</a></li>
</ul>
</aside>
</section>
@@ -116,7 +116,7 @@
<div class="tsd-signature tsd-kind-icon">Float<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L238">ctypes.ts:238</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L238">ctypes.ts:238</a></li>
</ul>
</aside>
</section>
@@ -126,7 +126,7 @@
<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</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/3f56a95b8/web/src/ctypes.ts#L236">ctypes.ts:236</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L236">ctypes.ts:236</a></li>
</ul>
</aside>
</section>
@@ -136,7 +136,7 @@
<div class="tsd-signature tsd-kind-icon">Null<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L240">ctypes.ts:240</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L240">ctypes.ts:240</a></li>
</ul>
</aside>
</section>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">TVMBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 12</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L248">ctypes.ts:248</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L248">ctypes.ts:248</a></li>
</ul>
</aside>
</section>
@@ -156,7 +156,7 @@
<div class="tsd-signature tsd-kind-icon">TVMDLTensor<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 7</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L243">ctypes.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L243">ctypes.ts:243</a></li>
</ul>
</aside>
</section>
@@ -166,7 +166,7 @@
<div class="tsd-signature tsd-kind-icon">TVMData<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 5</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L241">ctypes.ts:241</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L241">ctypes.ts:241</a></li>
</ul>
</aside>
</section>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">TVMModule<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 9</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L245">ctypes.ts:245</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L245">ctypes.ts:245</a></li>
</ul>
</aside>
</section>
@@ -186,7 +186,7 @@
<div class="tsd-signature tsd-kind-icon">TVMNDArray<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 13</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L249">ctypes.ts:249</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L249">ctypes.ts:249</a></li>
</ul>
</aside>
</section>
@@ -196,7 +196,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObject<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L244">ctypes.ts:244</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L244">ctypes.ts:244</a></li>
</ul>
</aside>
</section>
@@ -206,7 +206,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObjectRValue<wbr>Ref<wbr>Arg<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 14</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L250">ctypes.ts:250</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L250">ctypes.ts:250</a></li>
</ul>
</aside>
</section>
@@ -216,7 +216,7 @@
<div class="tsd-signature tsd-kind-icon">TVMOpaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L239">ctypes.ts:239</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L239">ctypes.ts:239</a></li>
</ul>
</aside>
</section>
@@ -226,7 +226,7 @@
<div class="tsd-signature tsd-kind-icon">TVMPacked<wbr>Func<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 10</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L246">ctypes.ts:246</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L246">ctypes.ts:246</a></li>
</ul>
</aside>
</section>
@@ -236,7 +236,7 @@
<div class="tsd-signature tsd-kind-icon">TVMStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 11</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L247">ctypes.ts:247</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L247">ctypes.ts:247</a></li>
</ul>
</aside>
</section>
@@ -246,7 +246,7 @@
<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L237">ctypes.ts:237</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L237">ctypes.ts:237</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/aynccallbackcode.html b/docs/reference/api/typedoc/enums/aynccallbackcode.html
index 667e1cddae..d929abf7c8 100644
--- a/docs/reference/api/typedoc/enums/aynccallbackcode.html
+++ b/docs/reference/api/typedoc/enums/aynccallbackcode.html
@@ -93,7 +93,7 @@
<div class="tsd-signature tsd-kind-icon">k<wbr>Exception<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 5</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L812">runtime.ts:812</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L812">runtime.ts:812</a></li>
</ul>
</aside>
</section>
@@ -103,7 +103,7 @@
<div class="tsd-signature tsd-kind-icon">k<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L811">runtime.ts:811</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L811">runtime.ts:811</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/dldatatypecode.html b/docs/reference/api/typedoc/enums/dldatatypecode.html
index 37a3cd9204..461ff46152 100644
--- a/docs/reference/api/typedoc/enums/dldatatypecode.html
+++ b/docs/reference/api/typedoc/enums/dldatatypecode.html
@@ -95,7 +95,7 @@
<div class="tsd-signature tsd-kind-icon">Float<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L339">runtime.ts:339</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L339">runtime.ts:339</a></li>
</ul>
</aside>
</section>
@@ -105,7 +105,7 @@
<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</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/3f56a95b8/web/src/runtime.ts#L337">runtime.ts:337</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L337">runtime.ts:337</a></li>
</ul>
</aside>
</section>
@@ -115,7 +115,7 @@
<div class="tsd-signature tsd-kind-icon">Opaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L340">runtime.ts:340</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L340">runtime.ts:340</a></li>
</ul>
</aside>
</section>
@@ -125,7 +125,7 @@
<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L338">runtime.ts:338</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L338">runtime.ts:338</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/rpcserverstate.html b/docs/reference/api/typedoc/enums/rpcserverstate.html
index 4e9afed8f8..6c785d4e86 100644
--- a/docs/reference/api/typedoc/enums/rpcserverstate.html
+++ b/docs/reference/api/typedoc/enums/rpcserverstate.html
@@ -90,7 +90,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Header<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
</ul>
</aside>
</section>
@@ -100,7 +100,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Header<wbr>Key<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
</ul>
</aside>
</section>
@@ -110,7 +110,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Server<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
</ul>
</aside>
</section>
@@ -120,7 +120,7 @@
<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Body<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L34">rpc_server.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L34">rpc_server.ts:34</a></li>
</ul>
</aside>
</section>
@@ -130,7 +130,7 @@
<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Header<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L33">rpc_server.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L33">rpc_server.ts:33</a></li>
</ul>
</aside>
</section>
@@ -140,7 +140,7 @@
<div class="tsd-signature tsd-kind-icon">Wait<wbr>For<wbr>Callback<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/sizeof.html b/docs/reference/api/typedoc/enums/sizeof.html
index 2fb5fb96bb..adbec9b89e 100644
--- a/docs/reference/api/typedoc/enums/sizeof.html
+++ b/docs/reference/api/typedoc/enums/sizeof.html
@@ -100,7 +100,7 @@
<div class="tsd-signature tsd-kind-icon">DLData<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = I32</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
</ul>
</aside>
</section>
@@ -110,7 +110,7 @@
<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = I32 + I32</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L229">ctypes.ts:229</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L229">ctypes.ts:229</a></li>
</ul>
</aside>
</section>
@@ -120,7 +120,7 @@
<div class="tsd-signature tsd-kind-icon">F32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
</ul>
</aside>
</section>
@@ -130,7 +130,7 @@
<div class="tsd-signature tsd-kind-icon">F64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
</ul>
</aside>
</section>
@@ -140,7 +140,7 @@
<div class="tsd-signature tsd-kind-icon">I32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
</ul>
</aside>
</section>
@@ -150,7 +150,7 @@
<div class="tsd-signature tsd-kind-icon">I64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
</ul>
</aside>
</section>
@@ -160,7 +160,7 @@
<div class="tsd-signature tsd-kind-icon">TVMValue<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
</ul>
</aside>
</section>
@@ -170,7 +170,7 @@
<div class="tsd-signature tsd-kind-icon">U16<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
</ul>
</aside>
</section>
@@ -180,7 +180,7 @@
<div class="tsd-signature tsd-kind-icon">U8<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/index.html b/docs/reference/api/typedoc/index.html
index 71189b2348..ccb0ebee5a 100644
--- a/docs/reference/api/typedoc/index.html
+++ b/docs/reference/api/typedoc/index.html
@@ -182,7 +182,7 @@
<div class="tsd-signature tsd-kind-icon">FObject<wbr>Constructor<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, lib<span class="tsd-signature-symbol">: </span><a href="classes/ffilibrary.html" class="tsd-signature-type">FFILibrary</a>, ctx<span class="tsd-signature-symbol">: </span><a href="classes/runtimecontext.html" class="t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L778">runtime.ts:778</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L778">runtime.ts:778</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Alloc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>shape<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, ndim<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeCode<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeBits<span class="tsd [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L113">ctypes.ts:113</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L113">ctypes.ts:113</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -288,7 +288,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>Bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">num [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L129">ctypes.ts:129</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L129">ctypes.ts:129</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -332,7 +332,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>To<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>from<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, to<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-sig [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L145">ctypes.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L145">ctypes.ts:145</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -376,7 +376,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>ToBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</sp [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L137">ctypes.ts:137</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L137">ctypes.ts:137</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -420,7 +420,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><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/3f56a95b8/web/src/ctypes.ts#L122">ctypes.ts:122</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L122">ctypes.ts:122</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -456,7 +456,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMBackend<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number< [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L161">ctypes.ts:161</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L161">ctypes.ts:161</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -508,7 +508,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMCFunc<wbr>Set<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ret<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L78">ctypes.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L78">ctypes.ts:78</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -556,7 +556,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMCb<wbr>Arg<wbr>ToReturn<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, code<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span c [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L84">ctypes.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L84">ctypes.ts:84</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -595,7 +595,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Call<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L68">ctypes.ts:68</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L68">ctypes.ts:68</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -651,7 +651,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><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/3f56a95b8/web/src/ctypes.ts#L58">ctypes.ts:58</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L58">ctypes.ts:58</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -687,7 +687,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Get<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span cla [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L101">ctypes.ts:101</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L101">ctypes.ts:101</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -726,7 +726,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>List<wbr>Global<wbr>Names<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>outSize<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, outArray<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L89">ctypes.ts:89</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L89">ctypes.ts:89</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -765,7 +765,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Register<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, f<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, override<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</spa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L95">ctypes.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L95">ctypes.ts:95</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -808,7 +808,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMGet<wbr>Last<wbr>Error<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -838,7 +838,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><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/3f56a95b8/web/src/ctypes.ts#L53">ctypes.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L53">ctypes.ts:53</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -874,7 +874,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Get<wbr>Function<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, funcName<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, queryImports<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">numbe [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -922,7 +922,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Import<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, dep<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-si [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -962,7 +962,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>obj<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><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/3f56a95b8/web/src/ctypes.ts#L169">ctypes.ts:169</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L169">ctypes.ts:169</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -998,7 +998,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Get<wbr>Type<wbr>Index<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>obj<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out_tindex<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L174">ctypes.ts:174</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L174">ctypes.ts:174</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1037,7 +1037,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Type<wbr>Index2<wbr>Key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>type_index<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, out_type_key<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><spa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1076,7 +1076,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Type<wbr>Key2<wbr>Index<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>type_key<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out_tindex<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L184">ctypes.ts:184</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L184">ctypes.ts:184</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1115,7 +1115,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMSynchronize<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>deviceType<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, deviceId<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signatur [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L151">ctypes.ts:151</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L151">ctypes.ts:151</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1157,7 +1157,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Alloc<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>size<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><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L189">ctypes.ts:189</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L189">ctypes.ts:189</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1193,7 +1193,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Free<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ptr<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><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/3f56a95b8/web/src/ctypes.ts#L192">ctypes.ts:192</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L192">ctypes.ts:192</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1229,7 +1229,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Func<wbr>Create<wbr>FromCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resource<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L209">ctypes.ts:209</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L209">ctypes.ts:209</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1269,7 +1269,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>args<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1321,7 +1321,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<wbr>Finalizer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resourceHandle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><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/3f56a95b8/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1357,7 +1357,7 @@
<div class="tsd-signature tsd-kind-icon">GPUPointer<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/3f56a95b8/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1372,7 +1372,7 @@
<div class="tsd-signature tsd-kind-icon">Packed<wbr>Func<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">...</span>args<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol"> & </span><a href="interfaces/disp [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L37">runtime.ts:37</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L37">runtime.ts:37</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1387,7 +1387,7 @@
<div class="tsd-signature tsd-kind-icon">Pointer<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/3f56a95b8/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1402,7 +1402,7 @@
<div class="tsd-signature tsd-kind-icon">Ptr<wbr>Offset<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/3f56a95b8/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1417,7 +1417,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObject<wbr>Base<span class="tsd-signature-symbol">:</span> <a href="classes/tvmobject.html" class="tsd-signature-type">TVMObject</a><span class="tsd-signature-symbol"> | </span><a href="classes/ndarray.html" class="tsd-signature-type">NDArray</a><span class="tsd-signature-symbol"> | </span><a href="classes/module.html" class="tsd-signature-type">Module</a><span class="tsd-signature-symbol"> | </span><a href="index.html#packedfunc" class="t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L781">runtime.ts:781</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L781">runtime.ts:781</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1435,7 +1435,7 @@
<div class="tsd-signature tsd-kind-icon">RPC_<wbr>MAGIC<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">1045105</span><span class="tsd-signature-symbol"> = 1045105</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/rpc_server.ts#L38">rpc_server.ts:38</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/rpc_server.ts#L38">rpc_server.ts:38</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1457,7 +1457,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/support.ts#L25">support.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/support.ts#L25">support.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1489,7 +1489,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/support.ts#L39">support.ts:39</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/support.ts#L39">support.ts:39</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1518,7 +1518,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/support.ts#L52">support.ts:52</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/support.ts#L52">support.ts:52</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1555,7 +1555,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/compact.ts#L38">compact.ts:38</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/compact.ts#L38">compact.ts:38</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1586,7 +1586,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1608,7 +1608,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/environment.ts#L32">environment.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/environment.ts#L32">environment.ts:32</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1639,7 +1639,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/compact.ts#L24">compact.ts:24</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/compact.ts#L24">compact.ts:24</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1661,7 +1661,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L1749">runtime.ts:1749</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L1749">runtime.ts:1749</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1726,7 +1726,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/support.ts#L62">support.ts:62</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/support.ts#L62">support.ts:62</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1748,7 +1748,7 @@
<div class="tsd-signature tsd-kind-icon">DLData<wbr>Type<wbr>Code<wbr>ToStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L343">runtime.ts:343</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L343">runtime.ts:343</a></li>
</ul>
</aside>
<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1757,7 +1757,7 @@
<div class="tsd-signature tsd-kind-icon">0<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "int"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L344">runtime.ts:344</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L344">runtime.ts:344</a></li>
</ul>
</aside>
</section>
@@ -1767,7 +1767,7 @@
<div class="tsd-signature tsd-kind-icon">1<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "uint"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L345">runtime.ts:345</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L345">runtime.ts:345</a></li>
</ul>
</aside>
</section>
@@ -1777,7 +1777,7 @@
<div class="tsd-signature tsd-kind-icon">2<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "float"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L346">runtime.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L346">runtime.ts:346</a></li>
</ul>
</aside>
</section>
@@ -1787,7 +1787,7 @@
<div class="tsd-signature tsd-kind-icon">3<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "handle"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L347">runtime.ts:347</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L347">runtime.ts:347</a></li>
</ul>
</aside>
</section>
@@ -1798,7 +1798,7 @@
<div class="tsd-signature tsd-kind-icon">Device<wbr>Enum<wbr>ToStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L272">runtime.ts:272</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L272">runtime.ts:272</a></li>
</ul>
</aside>
<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1807,7 +1807,7 @@
<div class="tsd-signature tsd-kind-icon">1<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "cpu"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L273">runtime.ts:273</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L273">runtime.ts:273</a></li>
</ul>
</aside>
</section>
@@ -1817,7 +1817,7 @@
<div class="tsd-signature tsd-kind-icon">15<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "webgpu"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L277">runtime.ts:277</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L277">runtime.ts:277</a></li>
</ul>
</aside>
</section>
@@ -1827,7 +1827,7 @@
<div class="tsd-signature tsd-kind-icon">2<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "cuda"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L274">runtime.ts:274</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L274">runtime.ts:274</a></li>
</ul>
</aside>
</section>
@@ -1837,7 +1837,7 @@
<div class="tsd-signature tsd-kind-icon">4<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "opencl"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L275">runtime.ts:275</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L275">runtime.ts:275</a></li>
</ul>
</aside>
</section>
@@ -1847,7 +1847,7 @@
<div class="tsd-signature tsd-kind-icon">8<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "metal"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L276">runtime.ts:276</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L276">runtime.ts:276</a></li>
</ul>
</aside>
</section>
@@ -1858,7 +1858,7 @@
<div class="tsd-signature tsd-kind-icon">Device<wbr>Str<wbr>ToEnum<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L280">runtime.ts:280</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L280">runtime.ts:280</a></li>
</ul>
</aside>
<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1867,7 +1867,7 @@
<div class="tsd-signature tsd-kind-icon">cl<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L283">runtime.ts:283</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L283">runtime.ts:283</a></li>
</ul>
</aside>
</section>
@@ -1877,7 +1877,7 @@
<div class="tsd-signature tsd-kind-icon">cpu<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L281">runtime.ts:281</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L281">runtime.ts:281</a></li>
</ul>
</aside>
</section>
@@ -1887,7 +1887,7 @@
<div class="tsd-signature tsd-kind-icon">cuda<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L282">runtime.ts:282</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L282">runtime.ts:282</a></li>
</ul>
</aside>
</section>
@@ -1897,7 +1897,7 @@
<div class="tsd-signature tsd-kind-icon">metal<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L286">runtime.ts:286</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L286">runtime.ts:286</a></li>
</ul>
</aside>
</section>
@@ -1907,7 +1907,7 @@
<div class="tsd-signature tsd-kind-icon">opencl<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L284">runtime.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L284">runtime.ts:284</a></li>
</ul>
</aside>
</section>
@@ -1917,7 +1917,7 @@
<div class="tsd-signature tsd-kind-icon">vulkan<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 7</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L285">runtime.ts:285</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L285">runtime.ts:285</a></li>
</ul>
</aside>
</section>
@@ -1927,7 +1927,7 @@
<div class="tsd-signature tsd-kind-icon">webgpu<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 15</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/runtime.ts#L287">runtime.ts:287</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/runtime.ts#L287">runtime.ts:287</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/interfaces/disposable.html b/docs/reference/api/typedoc/interfaces/disposable.html
index 83b2c69c20..42b82407a4 100644
--- a/docs/reference/api/typedoc/interfaces/disposable.html
+++ b/docs/reference/api/typedoc/interfaces/disposable.html
@@ -115,7 +115,7 @@
<div class="tsd-signature tsd-kind-icon">dispose<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/types.ts#L52">types.ts:52</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/types.ts#L52">types.ts:52</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/interfaces/functioninfo.html b/docs/reference/api/typedoc/interfaces/functioninfo.html
index 9d92fe8eaa..10951afee1 100644
--- a/docs/reference/api/typedoc/interfaces/functioninfo.html
+++ b/docs/reference/api/typedoc/interfaces/functioninfo.html
@@ -95,7 +95,7 @@
<div class="tsd-signature tsd-kind-icon">arg_<wbr>types<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
</ul>
</aside>
</section>
@@ -105,7 +105,7 @@
<div class="tsd-signature tsd-kind-icon">launch_<wbr>param_<wbr>tags<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
</ul>
</aside>
</section>
@@ -115,7 +115,7 @@
<div class="tsd-signature tsd-kind-icon">name<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/interfaces/libraryprovider.html b/docs/reference/api/typedoc/interfaces/libraryprovider.html
index b85045a5f3..d659ebfa20 100644
--- a/docs/reference/api/typedoc/interfaces/libraryprovider.html
+++ b/docs/reference/api/typedoc/interfaces/libraryprovider.html
@@ -112,7 +112,7 @@
<div class="tsd-signature tsd-kind-icon">imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/3f56a95b8/web/src/types.ts#L34">types.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/types.ts#L34">types.ts:34</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -127,7 +127,7 @@
<div class="tsd-signature tsd-kind-icon">start<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>inst<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">Instance</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/3f56a95b8/web/src/types.ts#L39">types.ts:39</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/e5ae4347d/web/src/types.ts#L39">types.ts:39</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/searchindex.js b/docs/searchindex.js
index 4e5f14a6ed..28fefeebc3 100644
--- a/docs/searchindex.js
+++ b/docs/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["arch/benchmark","arch/convert_layout","arch/debugger","arch/device_target_interactions","arch/frontend/tensorflow","arch/hybrid_script","arch/index","arch/inferbound","arch/introduction_to_module_serialization","arch/microtvm_design","arch/microtvm_project_api","arch/model_library_format","arch/pass_infra","arch/relay_intro","arch/relay_op_strategy","arch/runtime","arch/runtimes/vulkan","arch/security","arch/virtual_machine","contribute/ci","contribute/code_gu [...]
\ No newline at end of file
+Search.setIndex({docnames:["arch/benchmark","arch/convert_layout","arch/debugger","arch/device_target_interactions","arch/frontend/tensorflow","arch/hybrid_script","arch/index","arch/inferbound","arch/introduction_to_module_serialization","arch/microtvm_design","arch/microtvm_project_api","arch/model_library_format","arch/pass_infra","arch/relay_intro","arch/relay_op_strategy","arch/runtime","arch/runtimes/vulkan","arch/security","arch/virtual_machine","contribute/ci","contribute/code_gu [...]
\ No newline at end of file
diff --git a/docs/topic/vta/tutorials/autotvm/sg_execution_times.html b/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
index cfa1e4c836..8e3944a9c3 100644
--- a/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:30.634</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
+<p><strong>00:33.229</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 82%" />
@@ -354,11 +354,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-relay-vta-py"><span class="std std-ref">Auto-tuning a convolutional network on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_vta.py</span></code>)</p></td>
-<td><p>00:30.628</p></td>
+<td><p>00:33.222</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_alu_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-alu-vta-py"><span class="std std-ref">Auto-tuning a ALU fused op on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_alu_vta.py</span></code>)</p></td>
-<td><p>00:00.006</p></td>
+<td><p>00:00.007</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/topic/vta/tutorials/frontend/deploy_classification.html b/docs/topic/vta/tutorials/frontend/deploy_classification.html
index 78e9a8581f..071e4ea9a8 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_classification.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_classification.html
@@ -588,7 +588,7 @@ and dense layer which will both be executed in fp32 on the CPU.</p></li>
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 32.81s!
+resnet18_v1 inference graph built in 36.54s!
</pre></div>
</div>
</div>
diff --git a/docs/topic/vta/tutorials/frontend/deploy_detection.html b/docs/topic/vta/tutorials/frontend/deploy_detection.html
index 961b41e29d..8d5a4455ee 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_detection.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_detection.html
@@ -606,7 +606,7 @@ and dense layer which will both be executed in fp32 on the CPU.</p></li>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
-yolov3-tiny inference graph built in 22.30s!
+yolov3-tiny inference graph built in 24.42s!
</pre></div>
</div>
</div>
diff --git a/docs/topic/vta/tutorials/frontend/sg_execution_times.html b/docs/topic/vta/tutorials/frontend/sg_execution_times.html
index 2c7bdfa5c0..d067d367cf 100644
--- a/docs/topic/vta/tutorials/frontend/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/frontend/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-frontend-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>01:39.454</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
+<p><strong>01:45.411</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -353,12 +353,12 @@
<col style="width: 7%" />
</colgroup>
<tbody>
-<tr class="row-odd"><td><p><a class="reference internal" href="deploy_detection.html#sphx-glr-topic-vta-tutorials-frontend-deploy-detection-py"><span class="std std-ref">Deploy Pretrained Vision Detection Model from Darknet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_detection.py</span></code>)</p></td>
-<td><p>00:49.770</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" href="deploy_classification.html#sphx-glr-topic-vta-tutorials-frontend-deploy-classification-py"><span class="std std-ref">Deploy Pretrained Vision Model from MxNet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_classification.py</span></code>)</p></td>
+<td><p>00:53.678</p></td>
<td><p>0.0 MB</p></td>
</tr>
-<tr class="row-even"><td><p><a class="reference internal" href="deploy_classification.html#sphx-glr-topic-vta-tutorials-frontend-deploy-classification-py"><span class="std std-ref">Deploy Pretrained Vision Model from MxNet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_classification.py</span></code>)</p></td>
-<td><p>00:49.683</p></td>
+<tr class="row-even"><td><p><a class="reference internal" href="deploy_detection.html#sphx-glr-topic-vta-tutorials-frontend-deploy-detection-py"><span class="std std-ref">Deploy Pretrained Vision Detection Model from Darknet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_detection.py</span></code>)</p></td>
+<td><p>00:51.733</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/topic/vta/tutorials/optimize/sg_execution_times.html b/docs/topic/vta/tutorials/optimize/sg_execution_times.html
index a2e3bc1c39..02648c82eb 100644
--- a/docs/topic/vta/tutorials/optimize/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/optimize/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-optimize-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:03.153</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
+<p><strong>00:03.242</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,11 +354,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="convolution_opt.html#sphx-glr-topic-vta-tutorials-optimize-convolution-opt-py"><span class="std std-ref">2D Convolution Optimization</span></a> (<code class="docutils literal notranslate"><span class="pre">convolution_opt.py</span></code>)</p></td>
-<td><p>00:02.672</p></td>
+<td><p>00:02.729</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="matrix_multiply_opt.html#sphx-glr-topic-vta-tutorials-optimize-matrix-multiply-opt-py"><span class="std std-ref">Matrix Multiply Blocking</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply_opt.py</span></code>)</p></td>
-<td><p>00:00.481</p></td>
+<td><p>00:00.513</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/topic/vta/tutorials/sg_execution_times.html b/docs/topic/vta/tutorials/sg_execution_times.html
index d683bf69c4..c890066ba7 100644
--- a/docs/topic/vta/tutorials/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:00.814</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
+<p><strong>00:00.884</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -354,11 +354,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="matrix_multiply.html#sphx-glr-topic-vta-tutorials-matrix-multiply-py"><span class="std std-ref">Simple Matrix Multiply</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply.py</span></code>)</p></td>
-<td><p>00:00.420</p></td>
+<td><p>00:00.457</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="vta_get_started.html#sphx-glr-topic-vta-tutorials-vta-get-started-py"><span class="std std-ref">Get Started with VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">vta_get_started.py</span></code>)</p></td>
-<td><p>00:00.394</p></td>
+<td><p>00:00.427</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/tutorial/auto_scheduler_matmul_x86.html b/docs/tutorial/auto_scheduler_matmul_x86.html
index 4140c7694a..da19088f44 100644
--- a/docs/tutorial/auto_scheduler_matmul_x86.html
+++ b/docs/tutorial/auto_scheduler_matmul_x86.html
@@ -574,7 +574,7 @@ class Module:
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 96.193 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 100.398 ms
</pre></div>
</div>
</div>
@@ -646,7 +646,7 @@ automatically optimize a matrix multiplication, without the need to specify a
search template. It ends a series of examples that starts from the Tensor
Expression (TE) language that demonstrates how TVM can optimize computational
operations.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 32.083 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 24.282 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-auto-scheduler-matmul-x86-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../_downloads/eac4389b114db015e95cb3cdf8b86b83/auto_scheduler_matmul_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">auto_scheduler_matmul_x86.py</span></code></a></p>
diff --git a/docs/tutorial/autotvm_matmul_x86.html b/docs/tutorial/autotvm_matmul_x86.html
index e18d2eac75..4ec2466dda 100644
--- a/docs/tutorial/autotvm_matmul_x86.html
+++ b/docs/tutorial/autotvm_matmul_x86.html
@@ -685,16 +685,16 @@ reduce variance, we take 5 measurements and average them.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>waiting for device...
device available
Get devices for measurement successfully!
-No: 1 GFLOPS: 8.86/8.86 result: MeasureResult(costs=(0.030295845000000005,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.111386775970459, timestamp=1679614657.2000778) [('tile_y', [-1, 16]), ('tile_x', [-1, 128])],None,74
-No: 2 GFLOPS: 10.48/10.48 result: MeasureResult(costs=(0.0256183964,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6739628314971924, timestamp=1679614657.8733046) [('tile_y', [-1, 4]), ('tile_x', [-1, 32])],None,52
-No: 3 GFLOPS: 10.65/10.65 result: MeasureResult(costs=(0.0252048924,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7194416522979736, timestamp=1679614659.845875) [('tile_y', [-1, 4]), ('tile_x', [-1, 128])],None,72
-No: 4 GFLOPS: 0.95/10.65 result: MeasureResult(costs=(0.2813616966,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.752430438995361, timestamp=1679614664.6092327) [('tile_y', [-1, 32]), ('tile_x', [-1, 2])],None,15
-No: 5 GFLOPS: 1.98/10.65 result: MeasureResult(costs=(0.13565037119999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.4148261547088623, timestamp=1679614667.168229) [('tile_y', [-1, 8]), ('tile_x', [-1, 1])],None,3
-No: 6 GFLOPS: 9.84/10.65 result: MeasureResult(costs=(0.027278074199999995,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6956429481506348, timestamp=1679614669.171133) [('tile_y', [-1, 2]), ('tile_x', [-1, 256])],None,81
-No: 7 GFLOPS: 11.26/11.26 result: MeasureResult(costs=(0.023835292799999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6453514099121094, timestamp=1679614671.1168091) [('tile_y', [-1, 128]), ('tile_x', [-1, 32])],None,57
-No: 8 GFLOPS: 3.39/11.26 result: MeasureResult(costs=(0.07919194,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5253140926361084, timestamp=1679614672.6409886) [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
-No: 9 GFLOPS: 4.39/11.26 result: MeasureResult(costs=(0.061089533200000004,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2082717418670654, timestamp=1679614674.0398715) [('tile_y', [-1, 16]), ('tile_x', [-1, 16])],None,44
-No: 10 GFLOPS: 12.52/12.52 result: MeasureResult(costs=(0.021447108,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6481060981750488, timestamp=1679614674.6472747) [('tile_y', [-1, 256]), ('tile_x', [-1, 128])],None,78
+No: 1 GFLOPS: 10.03/10.03 result: MeasureResult(costs=(0.026752570600000004,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7407488822937012, timestamp=1679625052.3152878) [('tile_y', [-1, 16]), ('tile_x', [-1, 32])],None,54
+No: 2 GFLOPS: 11.13/11.13 result: MeasureResult(costs=(0.0241221214,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.649803638458252, timestamp=1679625052.971663) [('tile_y', [-1, 128]), ('tile_x', [-1, 512])],None,97
+No: 3 GFLOPS: 11.54/11.54 result: MeasureResult(costs=(0.0232565404,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6532163619995117, timestamp=1679625054.9199133) [('tile_y', [-1, 8]), ('tile_x', [-1, 256])],None,83
+No: 4 GFLOPS: 8.93/11.54 result: MeasureResult(costs=(0.0300584328,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7522788047790527, timestamp=1679625056.9847775) [('tile_y', [-1, 256]), ('tile_x', [-1, 256])],None,88
+No: 5 GFLOPS: 12.50/12.50 result: MeasureResult(costs=(0.02148152,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6188480854034424, timestamp=1679625057.795037) [('tile_y', [-1, 32]), ('tile_x', [-1, 128])],None,75
+No: 6 GFLOPS: 7.39/12.50 result: MeasureResult(costs=(0.0363478626,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.9304215908050537, timestamp=1679625059.9495907) [('tile_y', [-1, 1]), ('tile_x', [-1, 64])],None,60
+No: 7 GFLOPS: 0.51/12.50 result: MeasureResult(costs=(0.5292683636,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.71769094467163, timestamp=1679625068.6794431) [('tile_y', [-1, 128]), ('tile_x', [-1, 1])],None,7
+No: 8 GFLOPS: 2.10/12.50 result: MeasureResult(costs=(0.1277711432,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.311182975769043, timestamp=1679625070.9895515) [('tile_y', [-1, 128]), ('tile_x', [-1, 4])],None,27
+No: 9 GFLOPS: 10.56/12.50 result: MeasureResult(costs=(0.025417857,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6243619918823242, timestamp=1679625071.7493625) [('tile_y', [-1, 2]), ('tile_x', [-1, 512])],None,91
+No: 10 GFLOPS: 11.04/12.50 result: MeasureResult(costs=(0.024324382999999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6199319362640381, timestamp=1679625072.4000835) [('tile_y', [-1, 128]), ('tile_x', [-1, 32])],None,57
</pre></div>
</div>
<p>With tuning completed, we can choose the configuration from the log file that
diff --git a/docs/tutorial/autotvm_relay_x86.html b/docs/tutorial/autotvm_relay_x86.html
index d775137841..10352f90a6 100644
--- a/docs/tutorial/autotvm_relay_x86.html
+++ b/docs/tutorial/autotvm_relay_x86.html
@@ -563,7 +563,7 @@ standard deviation.</p>
<span class="nb">print</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">unoptimized</span></a><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{'mean': 513.497117569998, 'median': 513.9784247000023, 'std': 2.379501423445631}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{'mean': 512.8374819899979, 'median': 513.379893900003, 'std': 2.048731521479577}
</pre></div>
</div>
</div>
@@ -715,179 +715,177 @@ depending on the specifics of the model and the target platform.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 1/25] Current/Best: 6.77/ 15.66 GFLOPS | Progress: (4/20) | 12.27 s
-[Task 1/25] Current/Best: 18.37/ 18.37 GFLOPS | Progress: (8/20) | 15.88 s
-[Task 1/25] Current/Best: 22.95/ 23.24 GFLOPS | Progress: (12/20) | 19.64 s
-[Task 1/25] Current/Best: 11.16/ 23.24 GFLOPS | Progress: (16/20) | 22.56 s
-[Task 1/25] Current/Best: 3.16/ 23.24 GFLOPS | Progress: (20/20) | 25.84 s Done.
+[Task 1/25] Current/Best: 12.64/ 17.36 GFLOPS | Progress: (4/20) | 12.33 s
+[Task 1/25] Current/Best: 15.05/ 17.36 GFLOPS | Progress: (8/20) | 16.51 s
... 553 lines suppressed ...